Newly elected Democratic governors in Georgia, Ohio, and Texas have promised moratoriums on new data center construction and all took tougher stances on regulation than their incumbent Republican opponents. The trend carried over in various congressional races, in which candidates who were critical of their competitors’ record of promoting data center development came out on top, even flipping key districts in Ohio and South Carolina. 

Related

The AI Doomsday Future Is Not Inevitable

Many local communities, mostly in Ohio and Illinois, voted directly on their AI concerns and passed a flurry of ballot initiatives restricting data center development — close to 40 proposals, according to a tracker by research firm Redbud Consulting.

“What’s good for the AI industry is clearly not good for average voters,” said Shobita Parthasarathy, a public policy professor at the University of Michigan. “I wish that data center operators and the AI industry really take this as a sign that they’re done a very poor job of considering public concern and taking it seriously.”

Parthasarathy pointed out that many Americans are now deciding how they feel about these energy-guzzling facilities based on their experience of living near one and “real knowledge” about their deficiencies. Those concerns range from negative impacts on energy bills, to the environmental and health issues, but also include anger toward the outsized power that the tech oligarchy wields. 

Americans’ vote in the midterms has given them a tangible way to fight back, Parthasarathy said: Blocking the brick-and-mortar, water-sucking, heat-spewing buildings that power tech billionaires’ rapid wealth accumulation.

Tech Money ≠ Change Minds

A long track record of polls have recorded Americans’ anger about data centers. About three-quarters of voters polled by NBC News on Election Day said they opposed building these facilities in their communities. Those sentiments are consistent with a Heatmap survey from August that found a similar proportion of Americans would strongly oppose a new data center near where they live. 

More than half the voters polled by NBC this week said AI and data centers were a crucial or major factor in their decision, though they were almost equally divided when it came to deciding which party would be better at addressing their AI concerns. 

The AI industry tried to make a dent in the midterms, but much of their money was focused on a congressional primary in New York, where the cash helped defeat AI critic Alex Bores. 

Related

AI Tried to Buy the Midterms — Then the Lobby Learned Everyone Hates AI

In the general election, an Intercept analysis found, AI super PACs dramatically scaled back their spending on congressional races compared to primaries amid growing anger toward data centers; only 5 percent of their direct expenditures went toward contested general election races. Industry observers told The Intercept that this suggested developers and tech CEOs have taken note of the electorate souring on AI and data centers.

That’s not to say that tech magnates, more broadly, have been stingy with their money. Elon Musk spent nearly $450 million ahead of the midterms, including more than $200 million in Texas, where his companies are headquartered. 

“If Elon Musk wants to continue to burn half a billion dollars to get no results, there are probably other ways he can do that,” said Mitch Jones, deputy director of Food & Water Action, which has advocated for regulations on data centers on the federal, state, and local level. “It shows when an issue is deeply and broadly felt by American people, no amount of money can change their minds.”

Musk’s money wasn’t enough for the races in Texas to go his way. Surprise upsets in the state’s Senate and gubernatorial races resulted in victories for, respectively, Democrats James Talarico and Gina Hinojosa. 

Talarico promised to end tax breaks for data center developers and minimum federal requirements to protect the grid, environment, jobs, and transparency. Hinojosa called for a moratorium on all data center approvals “until laws can be passed by the Legislature to protect Texans.” Outgoing Republican Texas Gov. Greg Abbott had already temporarily halted environmental permitting and grid connections until about the end of the year, but Hinojosa’s proposal would last longer and be more expansive.

Promises to Rein In AI

Congress’s new incoming class includes many Democrats who have promised to rein in the harms of AI. Ohio’s Democratic Sen.-elect Sherrod Brown flipped a district after portraying his opponent in ads as cheerleading the data center industry.

In typically red South Carolina, Democratic Sen.-elect Annie Andrews defeated her opponent, Sen. Darline Graham, after accusing her of being bought by developers. Pro-AI super PAC Building the Future spent $2.5 million to support Graham, according to Politico. 

Related

Lesson From Lansing: To Win in Swing Districts, Fight Data Centers

In Mississippi, Democrat Scott Colom, who called for a temporary ban on data centers, was narrowly ahead of his incumbent opponent on Wednesday evening in a state that has not elected a Democrat to the Senate in more than four decades. 

Just because Trump is expected to veto Democrats’ proposals to regulate AI, that doesn’t mean they shouldn’t push for them, said Jones, of Food & Water Action. He’s hoping Congress will approve a ban on “artificial superintelligence,” as well as a federal moratorium on new data centers. 

“They should show the American public they are standing with us and taking actions to force Trump to veto those bills,” Jones said.

Although moratoriums typically target new construction and not existing data centers, Jones says they are still necessary to prevent an even more expansive buildout. 

“We don’t need any more,” he said, citing a Forbes estimate that the total cost of the projected buildout ranges from $2.8 trillion by 2030 to $10.30 trillion by 2032 in the U.S.

With Trump still in charge, states led by Democratic governors may be where more forceful regulation on data centers occurs. 

In Georgia, Democratic Gov.-elect Keisha Lance Bottoms promised a temporary statewide pause on data centers. Ohio’s new incoming Democratic Gov.-elect Amy Acton proposed an “immediate moratorium upon taking office on any new construction.” And in Kansas, Democrat Cindy Holscher, who declared victory in the gubernatorial race, has advocated for halting new development.  

Ballot Measures

Local communities are getting creative, too. Jones said that some towns and cities have turned to ballot measures to bypass “elected officials who are not keeping pace with their voters.”

Related

Facing Outraged Residents, Flock Got Help From Group That Uses AI to Rally Public Support

About half of the roughly 40 ballot measures in the midterms were in Ohio. Most succeeded by a wide margin, and as of Thursday morning, all but one vote were in favor of restrictions, according to Redbud Consulting’s tracker.

Nick Owens, who helped draft 10 local ballot measures in Ohio that restrict data center development after an unsuccessful attempt at a statewide petition, said this gave voters quicker and easier access to the ballot to curtail the “humongous growth of hyperscale data centers.”

“Communities know how best they want to govern themselves,” said Owens, who was a Trump delegate at the 2024 Republican National Convention and is currently co-founder of grassroots group Conserve Ohio. The results “just validated everything I’ve been seeing on the ground, talking to people, and in nationwide polling,” he added. 

Owens is looking ahead, once again, toward pushing for a statewide moratorium, as well as curtailing the secretive use of non-disclosure agreements by tech companies and developers. 

Parthasarathy, the public policy professor, said more communities may explore ballot measures as a response to developers’ lack of transparency and drive to build more. 

“I don’t think their thirst for data centers is going to go away,” she said. “They might move into localities where there is less organized pushback.”

The post Blue Tsunami’s Clearest Takeaway: People Hate Data Centers appeared first on The Intercept.

On Wednesday afternoon, The Associated Press declared Will Lawrence the winner of Michigan’s 7th Congressional District race over Republican Rep. Tom Barrett.

A national swing in favor of Democrats helped him — but so did a campaign that leaned heavily on opposition to data centers to
situate Lawrence as a populist instead of an extremist.

Data centers were on voters’ minds everywhere, but they rarely dominated a race as much as Michigan’s 7th District. Lawrence vocally opposed data center construction almost as soon as he began running last year. A series of proposed hyperscale developments in and around Lansing pushed the issue into the limelight.

A month and a half before the election, Lawrence announced a data center platform calling for a moratorium on new development, environmental regulations, and a national jobs program to replace work taken away by artificial intelligence.

Related

Abdul El-Sayed Wins Senate Seat With Pro-Palestine Campaign

As progressives take stock of the 2026 midterms, Lawrence’s approach points one way forward for their wing of the Democratic Party. Lawrence’s opponent tried to turn his time as an environmental activist into a liability, but in an interview with The Intercept, Lawrence said his background primed him to come down on the same side as the public on a red-hot issue like data centers.

“I fought against Wall Street landlords trying to kick people out of their homes. I fought against oil companies who are polluting our air and water,” he told The Intercept in an interview last month. “So for me, this was not the first time that I’ve stood on the line against a corporate behemoth like this.”

From Sunrise to Congress

Lawrence is a co-founder of the Sunrise Movement, which had one of its most famous moments when its youth activists marched into the office of then-House Speaker Nancy Pelosi to push for the Green New Deal with Rep. Alexandria Ocasio-Cortez, D-N.Y., by their side in 2018.

Data centers and AI were remote issues for voters then, but they were starting to bubble up in the public consciousness by the time Lawrence announced his run for Congress in August 2025.

Related

AI Tried to Buy the Midterms — Then the Lobby Learned Everyone Hates AI

Two months later, one of his first moves as a candidate was to visit public meetings in two small townships in Michigan to express his opposition to proposed data center developments.

As a former member of the Democratic Socialists of America, he drew skepticism from many observers over his ability to win the primary in a district that narrowly voted for Donald Trump in 2024. They were equally skeptical of his ability to beat Barrett, a Republican who tried to play up his moderate bona fides in the race’s final weeks. Barrett was one of a handful of Republicans to vote in favor of a resolution to end the Iran war, for instance.

Lawrence had some major liabilities in a purple district, including tweets where he wrote “fuck the nuclear-family home” and told then-Democratic Sen. Joe Manchin in a song to “hurry up and die.”

While Barrett leaned on those posts and Lawrence’s time with the DSA to cast his opponent as an extremist, the young progressive had a card up his sleeves in the form of the data center issue.

It is the rare cause that leaps across party lines to unite Republicans and Democrats who are just as worried about water sources drying up and ceaseless buzzing sounds.

Related

The AI Doomsday Future Is Not Inevitable

At an October 5 city council meeting in the quaint town of Mason, where a developer has proposed a hyperscale data center on a plot of picturesque farmland, Democrats and Republicans alike said they were outraged. Some residents said they would sell their house and leave town if the development is built.

In a sign of how politically charged the issue has become, many of the speakers were candidates for office angling for votes. Lawrence had dropped by himself a few weeks before. His first television ad centered on his opposition to the Mason project.

Karl Sime, a Mason resident who crisscrossed the area on behalf of Democratic candidates, told The Intercept last month that “100 percent” of voters were angry about the data center proposed for the swath of farmland just outside city limits.

“Sometimes they’ll just say it unprovoked,” he said. “You can see the body language just dramatically change into anger.”

Republican Countermoves

Lawrence was hardly the only candidate who tried to tap into opposition to data centers. Some Republicans tried to catch the wave as well.

Related

How a Small Kansas City Became a Critical Hub for Trump’s Deportation Regime Against Its Will

Barrett voted for the One Big Beautiful Bill, which included tax breaks that could benefit rural data center developers. But as the danger he faced in the election became clear, he proposed legislation in Congress that would ban local officials from signing non-disclosure agreements with developers and would allow local communities to set regulations on data centers without federal interference.

Some voters said they wanted to see candidates take a harder stance in favor of regulation, however. At the City Council meeting last month, Doug DeLind said he was left shaking his head at neighbors who had yard signs opposing data centers and supporting Barrett.

DeLind, a sculptor who lives near the proposed data center, said that Barrett’s position was that “we’re going to leave it up to the local communities, but we’re not going to provide any federal protection or any guidelines to what a data center can do.”

Although Mason faces a more imminent possibility of a massive data center, Americans across the country are worried about having one in their backyard. Multiple surveys showed a remarkable shift in public opinion on the subject over the past year. Half of Americans oppose a data center being built within 3 miles of their house, according to a recent Politico poll.

Related

Francesca Hong’s Loss in Wisconsin Is a Win for AI Data Centers — for Now

In recent months, data centers became a signature issue for Democrats with progressive pasts trying to recast themselves as populists. Sherrod Brown, the party’s Senate candidate in Ohio, and Texas gubernatorial candidate Gina Hinojosa also staked their campaigns on opposition to data centers. Both won.

“The playbook was listening to voters. I think voters across ideologies were speaking with one voice, and I think their opponents were some of the people most willing to silence voters and accept industry cash instead,” said Alex Bores, who became a national figure by arguing for stricter AI regulations during a failed New York congressional primary bid earlier this year.

As candidates ran away from data centers, the artificial intelligence industry’s most powerful super PAC stopped intervening in close races. An Intercept analysis found that the well-funded Leading the Future super PAC network, which is aligned with OpenAI, largely stopped spending in competitive general election races.

It lost in the one state where it did place a big bet, South Carolina. Incumbent Sen. Darline Graham, a Republican, fell to challenger Annie Andrews.

Bores, who now leads a pro-AI regulation group called Who Decides, said he was not surprised to see candidates increasingly market themselves as AI skeptics.

“I think it’s overdue. I think candidates and elected officials are catching up to where the electorate is, across parties,” Bores told The Intercept. “It is remarkable the agreement that voters have around wanting guardrails to protect us from the most powerful and fastest moving technology of our lifetime. Democracy is sometimes slow, but only so much.”

This story was supported by a grant from the Tarbell Center for AI Journalism.

Update: November 4, 2026, 2:32 p.m. ET
The story has been updated after The Associated Press officially called Michigan’s 7th Congressional District race for Will Lawrence.

The post Lesson From Lansing: To Win in Swing Districts, Fight Data Centers appeared first on The Intercept.

Super PAC networks bankrolled with millions of dollars from Anthropic and OpenAI executives spent heavily in competitive primaries earlier this year, including a high-profile race in New York City. They were touted as the looming successor to another single-issue group, the American Israel Public Affairs Committee, as AIPAC moved its spending underground.

Yet as public outrage over data centers and AI’s threat to jobs soared, artificial intelligence super PACs dramatically throttled back spending on congressional races compared to the primaries. The biggest AI super PAC network has spent only modestly ahead of next week’s general election — and mostly in safe seats.

An Intercept analysis of AI spending in the 2026 midterms, based on Federal Election Commission records and independent race ratings, shows that only 5 percent of their direct expenditures have gone to contested general election races. The industry’s avoidance of the contested races that could determine control of the next Congress suggest that the PACs have realized how the public has soured on the tech, campaign finance and industry observers told The Intercept.

Related

AIPAC Isn’t Retreating. Here’s How It’s Already Shaping the Next Congress.

“Essentially these companies underestimated how much the public despises the data center buildout, and the inescapability of AI systems that no one asked for that say they will automate jobs and maybe kill us all,” said Rick Claypool, a research director for the president’s office at Public Citizen, a nonprofit campaign finance watchdog.

Although there is still time for a last-minute flurry of spending, the story of AI’s role in the 2026 midterm elections may turn out to be the story of a single race — a primary in New York — that forced the industry to step away from the limelight.

Claypool’s group released an analysis earlier this year of “corporate supremacist” super PACs, a new breed of campaign committees modeled on a crypto super PAC called Fairshake that spent tens of millions on the 2024 elections.

The model was to amass a massive war chest and use it to punish candidates who opposed the company’s preferred government regulations, while throwing cash at allies or potential allies from both parties.

Related

Who’s Spending in Your Congressional Election? We Tracked the Front Groups Fueling the 2026 Midterms.

The first big AI super PAC to launch in August 2025 was Leading the Future, which was bankrolled by the president of OpenAI and its investors. A rival super PAC network, Public First, launched months later with the goal of countering Leading the Future’s influence. Its donors include Anthropic CEO Dario Amodei, who gave $1 million. Anthropic has also given a nonprofit associated with Public First millions of dollars, while stipulating that the money cannot be used on elections. Public First says that it supports tighter regulations than its rival.

The two networks plowed $50.7 million into the primaries, campaign finance records show. Most of that money was spent on primary races where the outcome was in doubt, driven by an eye-popping $21.3 million the networks spent on supporting or opposing a single congressional candidate, Alex Bores, in the Democratic primary for New York’s 12th Congressional District.

Bores, a New York State Assembly member, attracted the ire of those in the industry who favor a laissez-faire approach by authoring an AI regulation bill in the state legislature.

Leading the Future succeeded in defeating Bores, who lost to his State Assembly colleague Micah Lasher in June. But the group drew intense blowback in the process. Many observers suggested Leading the Future’s spending had boosted Bores rather than hurt him, because it gave him a chance to position himself as a foe of the AI industry — while also benefiting from a PAC aligned with the Public First network.

“Down to the wire, they were sweating,” Bores told The Intercept, referring to Leading the Future. “I think people who have been in politics for a while will realize that their spending really served to elevate the issue of AI.”

Leading the Future did not respond to a request for comment.

After wading into contested primary elections, the two leading AI super PAC networks have held back from spending on competitive general elections — even though they have plenty of money in the bank.

Until a few days ago, Public First had spent nothing on general election races. Leading the Future has spent only $7.4 million on general election races as of October 26, according to the FEC data, despite having $34 million in cash on hand across its network as of its most recent report.

Only 31 percent of that total has gone toward competitive races, as measured by the independent analytics website FiftyPlusOne. The only reason Leading the Future has spent money on competitive general election races at all is because it’s backing U.S. Sen. Darline Graham, R-S.C., who filled the seat of her late brother Lindsey Graham and is now facing an unexpectedly fierce challenge from a Democrat to win a full term in ruby-red South Carolina.

“They are not going to have an impact on the balance of power in Congress,” said Shaunna Thomas, one of the founders of Guardrails Alliance, another new super PAC set up to counter Leading the Future’s influence. “They are spending money on candidates who they think will protect their ability to control if and how they are regulated.”

Public First has only recently started spending a relatively small sum on a handful of general election races. The spending had amounted to $1.1 million as of the morning of October 26, according to processed FEC data.

Asked for comment on how its strategy has changed since the primaries, a Public First spokesperson said that it was still trying to support “strong advocates for AI regulation.”

“Public First was founded as a counterweight to Leading the Future, and that fight looks very different today. Leading the Future is going dark. Its biggest backers are walking away,” said Anthony Rivera-Rodriguez, the spokesperson.

Related

Google DeepMind Engineer Resigns in Protest of Pentagon Contract

Into that gap have stepped new groups pushing for tougher AI regulation such as the Guardrails Alliance, which bills itself as a grassroots alternative to corporate-backed super PACs. It is funded in large part by Silicon Valley tech workers but has taken a more aggressive stance against “tech oligarchs.” Guardrails has spent about $2 million thus far and reported having $1.2 million on hand as of October 14.

About three-quarters of voters are less likely to support a candidate they know has received support from an AI super PAC, according to a Data for Progress poll of more than 3,000 likely voters in Iowa, Michigan, and Ohio commissioned by Guardrails. Voters were more concerned about spending from AI companies than from crypto companies or AIPAC.

The dislike for data centers and AI cuts across party lines. In Michigan’s purple 7th Congressional District, Democratic candidate Will Lawrence has seized on data centers as one of his leading issues.

“The politics of this are terrible. Eighty percent of everybody is against it, whether you’re a Democrat, independent, or Republican,” he told supporters at a recent town hall.

Notably, neither of the super PAC networks has intervened in Lawrence’s race, despite it being one of the closest in the nation. (Lawrence’s campaign has pointed out that Elon Musk, who leads AI company xAI, has spent against Lawrence via his America PAC.)

There are emerging signs that the big AI super PAC networks are changing tactics in response to that public distaste. Public First’s dark-money affiliate donated $3 million since August to a preexisting super PAC supporting California state Sen. Scott Wiener, one of two Democrats vying to replace retiring Rep. Nancy Pelosi in California.

Leading the Future, meanwhile, gave a $500,000 donation on October 9 to a pro-data center nonprofit called Build American AI.

AI super PAC critics are also watching to see what Leading the Future will do, given its large war chest.

“I think they are going to continue to spend. I think they are going to try to do more and more spending that isn’t as easily tracked, because they are such political dead weight,” said Bores.

Delaware Corporations

When single-issue super PACs know their issue is unpopular, they use vague names, “pop-up” affiliates, and other dubious tactics to hide their activities from voters.

Add another one to the list: Leading the Future has routed more than $30 million of spending through newly minted Delaware corporations, blocking transparency during an election that has been defined by voter backlash against the industry.

Those apparent shell corporations have accounted for the lion’s share of the largest artificial intelligence super PAC network’s expenses during this year’s election.

The practice is a flagrant violation of Federal Election Commission guidelines that require campaign organizations to disclose which consultants, pollsters, and advertisers they are paying, a campaign watchdog alleged in a complaint filed in May.

There may be little recourse for observers worried about the lack of transparency for now, however. The FEC is short of the quorum that would allow it to take action on complaints such as the one filed by the Campaign Legal Center. (Leading the Future, which did not respond to a request for comment, has previously defended its tactics as legal in a statement to Wired.)

The Leading the Future network was funded in large part by an executive at or investors in OpenAI, one of the world’s largest AI companies. (In 2024, The Intercept sued OpenAI in federal court over the company’s use of copyrighted articles to train its chatbot ChatGPT. The case is ongoing.)

Company president Greg Brockman and his wife, Anna Brockman, donated $25 million to the network last year. When it came time to spend that money, the Leading the Future network leaned on two affiliate super PACs, one for Democratic candidates called Think Big and the other for Republican candidates called American Mission.

Related

AIPAC, AI, Crypto, and Gambling Are Hiding Their Big Election Spends

That is a typical tactic of the new breed of single-issue super PACs popularized by the crypto industry. Campaign finance reform advocates say it may help obscure the origin of campaign ads from voters but does not run afoul of any laws itself.

What those super PACs did next raised red flags for watchdogs, however. Since December, American Mission has reported spending $17.4 million through a Delaware-incorporated limited liability corporation called Summit Ridge Media Group, on services ranging from media placement to ad production to consulting.

Think Big has spent $21.3 million through a different Delaware corporation, Lantern Production Consultants, on a similarly wide range of expenses.

The grab bag of expenses is a giveaway that the super PACs are likely using the newly created corporations to hide their true vendors, according to Shanna Ports, a senior legal counsel at the Campaign Legal Center.

“There’s no way of knowing what is motivating this arrangement — but I know there is backlash to AI. It is possible that the vendors who are working with American Mission and Think Big don’t want to be identified in FEC reports,” Ports said.

Thomas, of the rival super PAC Guardrails Alliance, told The Intercept she thinks the network is obscuring its spending because a lax regulatory approach has become “politically toxic.”

“Their donors don’t want to be associated, their candidates don’t want their support, and their vendors don’t want to be associated,” Thomas said. “If you were public, you are running away. If you are not, you are interested in not being exposed.”

This story was supported by a grant from the Tarbell Center for AI Journalism.

The post AI Tried to Buy the Midterms — Then the Lobby Learned Everyone Hates AI appeared first on The Intercept.

When it was founded a year ago, news reports suggested that Fellowship PAC would concentrate on supporting Republicans.

Since June, however, the only candidates Fellowship PAC has spent on have been Democrats, including those who have criticized the president’s meme coin, and it is getting help from a former Trump administration official’s company.

Fellowship PAC funneled $600,000 since this summer through a firm co-founded by former Trump crypto adviser Bo Hines to help reelect Rep. Ritchie Torres of New York; Rep. Maggie Goodlander of New Hampshire; and Sen. Mark Warner of Virginia. The Democrats are considered crypto allies, and none of them are facing competitive elections this fall.

All those advertising dollars have been spent through a vendor called Nxum Group, which was co-founded by Hines, a former Republican congressional candidate and Trump administration official.

Hines did not return a request for comment, and it is not known whether he has personally been involved in the Nxum Group’s work on behalf of Fellowship PAC. In congressional races in 2022 and 2024, he presented himself as a Trump-loving MAGA acolyte.

There is a thick web of ties between Hines, Fellowship PAC, and one of the biggest players in the crypto industry, a stablecoin company called Tether.

Before Trump’s election, Hines was involved in promoting a Trump-themed meme coin that made a handful of investors millions before plummeting in value. After Trump’s election, the president tapped Hines to serve as the head of the White House Council of Advisers on Digital Assets. In that role, Hines helped pass a law that created stablecoin regulations that benefited Tether.

As Hines served in the Trump administration, Tether was paying money to the firm that Hines co-owned, Nxum Group, according to a report last week in the Wall Street Journal. Nxum Group told the news outlet that Hines had no involvement in the company’s business dealings with Tether during that time. Hines said in a June 2025 ethics disclosure that he had only a “passive investment interest” in the company.

Hines left the Trump administration in August 2025 and, a month later, Tether announced that he would serve as the CEO of its U.S. branch.

Related

Who’s Spending in Your Congressional Election? We Tracked the Front Groups Fueling the 2026 Midterms.

Tether also has ties to Fellowship PAC. The newly created Fellowship PAC received a $10 million donation in January from Tether’s U.S. banking firm, Cantor Fitzgerald. In April, Fellowship PAC announced that Tether’s vice president of regulatory affairs would serve as its chair.

Although it launched with a splash last year by claiming that it had a $100 million war chest, the PAC has never reported as much as that in official campaign finance disclosures. The PAC’s last receipt was in January, and its cash on hand dwindled to $250,000 as of October 14.

Throughout the midterm elections, Fellowship PAC has leaned heavily on Nxum Group to spend advertising dollars on various congressional races.

Fellowship PAC spent exclusively on Republicans until its first expenditure on behalf of a Democrat, Torres, in June. That spending coincided with public polls showing increasingly dismal fortunes for Republicans in the midterm elections. Since then, the PAC has spent exclusively on Democratic candidates.

Torres’s and Warner’s campaigns did not respond to requests for comment. Goodlander’s campaign said she had nothing to do with the $150,000 that Fellowship PAC spent on digital advertising in her race. The money, the PAC disclosed in filings, had been routed through Nxum Group on August 31.

“Congresswoman Goodlander has been one of the strongest advocates in Congress for cracking down on Big Tech and crypto profiteering and getting dark money out of politics,” her campaign said in a statement. “Like all campaigns, we’re legally barred from coordinating with outside groups like these, so you’d need to ask them why they’re spending in this race.”

The PAC did not return a request for comment.

Related

Crypto Darling Gillibrand’s 11th-Hour Flip-Flop Helps Block Trump Crypto Bill

Goodlander, Torres, and Warner are all rated as strong supporters of the crypto industry on an industry scorecard. Still, all three have been critical of Trump’s personal dealings in the industry though his $TRUMP meme coin, which has helped make him $1.4 billion from crypto ventures.

Warner recently ran an ad of his own attacking Trump over the meme coin.

“That’s not governing. It’s grifting,” the Democratic senator said. “The corruption needs to stop.”

The post Giant Crypto Super PAC Funnels Big Money to Former Trump Official’s Firm to Support… Democrats? appeared first on The Intercept.

Disenchanted and disturbed by Google’s military contracts with the U.S. and Israel, Liguori says she could no longer stand working at a company that was going to help kill people or help build technology that “automatically makes it easier to escalate the brutality of war.”

Related

The Pentagon Asked OpenAI for Artificial Intelligence Designed to Rarely Say No

Liguori joined DeepMind in 2023 and began working on AI systems aimed at more efficiently routing power through the electrical grid, a peacetime engineering problem. But as she learned of Google’s Project Nimbus contract, which furnishes Israel’s military and top weapons manufacturers with a suite of advanced artificial intelligence tools, she began to question her role in the company, and the company’s role in the world. The day after Google announced its classified military AI contract with the Pentagon in April, Liguori staged a personal sick-out, using a combination of sick leave and holiday time off to cease her work.

In a letter she sent internally informing DeepMind and her colleagues of her resignation, Liguori urged them to “stay human” in the face of military work. “I no longer believe that our technology will have a net positive impact on society. I think there’s a fantasy world where AI can help decarbonize grids, cut waste, protect ecosystems, and benefit humanity,” she wrote. “But it’s not this world, and it’s not this type of general purpose AI we are racing to develop. Racing to deploy agentic AI in this society and historical moment is a terrible idea, and will definitely be a net evil.”

The Intercept spoke to Liguori, an Italian national living in London, where DeepMind is headquartered, about how she came to view her employer as dangerously powerful. During a series of interviews by phone and via text, which have been edited and condensed for length and clarity, she explained her moral opposition to military work, the role of unionization in dissent, and what hope engineers and scientists might have for influencing their companies.

What was it about DeepMind that made you want to work there?

I had a high opinion of DeepMind; it was a dream company. It had this aura of being a research-oriented place where you could sit close not just to computer scientists, but real scientists and researchers and people with backgrounds in biology, in physics, in mathematics, in ecology, in anthropology.

You joined in 2023, by which point the company had already been involved in Project Nimbus. Were you aware of that project at the time?

Not really. The first year when I joined, I was very new and very isolated in my team. Google DeepMind is a huge company and if you’re not in the right channels, you end up being very siloed. I didn’t know about Project Nimbus, but when the Gaza war escalated, I discovered it. I immediately looked for some internal resistance movement to join, but I didn’t find anything.

I discovered [Google employee advocacy group] No Tech for Apartheid, but I thought it was mainly a U.S. organization. I didn’t know how to connect with colleagues that share concerns, so the only thing that I did was buy a Palestine hoodie and go to the office with that.

Related

Israeli Weapons Firms Required to Buy Cloud Services From Google and Amazon

In the summer of 2024, I went on maternity leave because my daughter was born, and during maternity leave I was watching news from Gaza every day. I was really suffering experiencing maternity and watching a livestreamed genocide, and every time I was thinking of my company’s complicity in that. I thought about not coming back to work at all.

When I was due to come back, I met a colleague that had basically taken my role in my team before maternity leave, and they were actually leaving for ethical reasons. They put me in contact with the union movement and No Tech for Apartheid. And so I found other colleagues fighting against this.

How and when did you learn of the company’s contracts with the U.S. military?

By the time they started engaging with the U.S. Department of War, [Editor’s note: July 2025] I was already involved in activism. By this winter we saw the GenAI.mil contract materializing, and for me, this was a very clear indication of the company’s intentions. At the same time Google ramped up the security level authorization for Cloud: IL6, which is the level required to handle classified workloads.

Related

AI Almost Started a U.S.–China War — and No One Seems to Care

The combination of the two — signing this non-classified deal with the Department of War, and ramping up security level — really meant that we wanted to use AI for war, on the battlefield. Even if the internal narrative was, “No, don’t worry, it’s just for administrative purposes, just to process documents.”

What was happening internally?

Internally there was a lot of conversation in specific channels where people were complaining, but leadership never really answered anything. I had some of my posts deleted.

What had you written, and did you ever get any sort of answer as to why this stuff was deleted?

I had several posts deleted because they contained the word “genocide,” and I was told that genocide is an intrinsically violent word, and so I was violating internal policies. Even when my posts weren’t deleted, I often had conversations with my manager and HR telling me that my tone was not appropriate because I was too aggressive and that I should raise my concerns in different ways.

They basically wanted us to send [internal HR] tickets to raise our concerns. The problem is that when you do that, you receive semi-automated answers with absolutely no content, nobody is engaged on the concern you have raised. There’s no discussion at all, you don’t even interact with a human. It’s a lot of administrative and bureaucratic steps you have to follow to raise concerns. What I wanted was discussion and what I wanted was very clear answers from leadership.

They tell you this is not the place to raise concerns, try to go through this other channel. This is the common technique. They always tell you to do something else, try to make you tired, and you will never get an answer.

[Editor’s note: The Intercept asked Google about Liguori’s criticism of both the contract and how discussion is handled internally. Spokesperson Tom Hewett responded: “We work with public sector organisations worldwide and follow a consistent process for reviewing and entering these contracts, including compliance with our acceptable use policies and AI principles. We also have dedicated channels for employees to safely raise any concerns – ranging from business ethics to misconduct – and have strict workplace policies in place to prevent retaliation of any kind.”]

I think the reason why you would be opposed to Project Nimbus is pretty straightforward: genocide. What were your specific concerns about or opposition to the Department of War contract?

I’m against the use of AI for targeting and in warfare in general, not only for autonomous weapons. I’m against AI for war, and I don’t think a commercial company like Google should have anything to do with that.

AI for targeting is very dangerous; it creates a system where it’s easier to escalate the brutality of war. The first problem is the opacity introduced in the definition of targets and the dehumanization that follows. In principle, if you have to define targets, if you want to define Hamas combatants, you have to think what it actually means and come up with a definition.

But with these systems, you can bypass the definition. The system itself comes out with a definition. You don’t even question it, you just blindly rely on the definition as the ultimate truth. The person just becomes a set of aggregated data points. All individuality is lost, all humanity is lost in in this process.

Related

Facing Years in Prison for Drone Leak, Daniel Hale Makes His Case Against U.S. Assassination Program

It becomes a system that dehumanizes people. It becomes a rubber stamp because the human brain cannot process all this information. You keep a person that formally approves it, but they don’t, they cannot, meaningfully evaluate the output. This links to the accountability gap: Everyone in the kill chain can plausibly say that they didn’t decide.

What did you do to protest the Department of War contract internally?

Mainly messaging on corporate channels to inform colleagues, and speaking with them in person. I tried to channel the widespread dissatisfaction and indignation over the contract into the collective movement we are building, into our union effort. [Editor’s note: DeepMind employees voted to unionize in May.] Many people still don’t know there are other colleagues organizing against these contracts and for a more ethical Google DeepMind.

Related

The AI Doomsday Future Is Not Inevitable

When you are alone and you lose faith in your company, and you stop believing your work makes a positive difference, you can only become disillusioned and frustrated, and you ultimately end in acceptance and cynicism, or you leave. Finding you are not alone, and joining a collective movement, gives a much better option. It gave me the strength to stay, at least for a year.

I also co-organized a demonstration in February in front of the office. The scope was broader than just the Department of War, but the December GenAI.mil launch had given me a deep sense of urgency, and that’s why I organized the demonstration. I knew something like the classified deal was coming, and I knew this meant legitimizing Nimbus-style collaborations to become the standard. The demonstration was for Gaza and for any use of AI in war. We had banners with “No AI in the kill chain” and “No Gemini for Genocide.”

Has your work as a research engineer contributed to your concerns about this technology being used to kill? How has your technical knowledge informed your opposition?

Understanding how these systems work makes it easier to see where the problem is. Understanding statistical aggregation, and how easy it is to just delegate responsibility, delegate decision-making, makes a difference. But I think it’s more important to do some philosophical reasoning here.

Do you have concerns about the reliability of the technology?

I am concerned about unreliability. But I don’t think you should ever look at unreliability separate from a discussion of why AI in warfare is wrong by itself.

A lot of the time what I hear is a focus on error, for example, the Minab school attack. If we focus only on the error, we are basically saying that if there was no error, these systems for automated targeting are OK.

Were there people within Deep Mind who were supportive of this contract or military work generally?

Yes, it was actually a very debated topic. I think there is broad consensus on the Anthropic red lines, which are fully autonomous weapons and mass surveillance.

Related

Pete Hegseth’s Military of the Future: Defined by AI and an Utter Lack of Oversight

But on whether to use AI for military purposes, this is a much more debated topic, and a lot of people buy the narrative of us-versus-them: That we need to defend ourselves against Russia, against China, and so on.

What was your thought process on deciding to leave the company over this?

It was a very challenging year. I came back as a new mom with a full-time job, a baby to handle, I had some health issues, and I was committing to activism basically on the side. Many people told me, you can slow down, but I couldn’t. I couldn’t stay in the company and just go on with business as usual. I really felt we needed to stop the Department of War contract because I perceive that contract as normalizing what happened in Gaza — the kind of tech-powered, AI-powered, brutal war that was tested in Gaza was getting approved to become the standard of war.

I don’t think petitions have a great effect in general. The [DeepMind] petition summarized my disillusionment from us not having done more during the year to prevent this from happening. For me, really, we should have gone on strike this year. We should have stopped working.

In the U.S., there is a lot of defense of these contracts on the basis that it’s important for national security. That might be persuasive if you’re American. But Americans are a pretty small minority on the planet. I’m curious what your perspective is on that kind of rationale as someone who lives in another country, who is a citizen of another country.

There’s been discussion about that as well. A lot of Europeans were actually scared of that, because the current Trump administration is not very friendly even to Europeans. What feels like national security for U.S., people feels like a threat for Europeans.

Are you aware of other colleagues at DeepMind who are going to follow what you’re doing and quit?

Some have quit already, some publicly and some silently. There have been many people leaving because of ethical reasons. Not all of them publicized it.

Do you think it’s possible for workers to change something like the Pentagon contract? What is the worker movement inside Google/DeepMind missing that would allow for that kind of change?

I think it’s possible, but it requires a real commitment and readiness to take some risk. Ultimately, I think we need a workforce ready to strike to oppose these contracts. Currently, many people are too scared to even express an opinion publicly. The company was very effective at creating a pervasive sense of fear, which induces self-censorship.

What I tried to show my colleagues this year was exactly that the boundaries we perceive and we retreat behind are not the real boundaries — that we can speak out and keep our jobs. Most of the times the reaction is intimidation in meetings, mentions of policy violations, but nothing else follows. Obviously, there are some boundaries, but they are much farther then the perceived ones.

I also took a day off and demonstrated in front of the office in February, together with University College London BDS [Boycott, Divestment, Sanctions movement] staff and a few other colleagues. I faced literally no consequences for that because it’s a trade union activity in my spare time, and in the U.K., we are protected for such an action. This can be a form of soft strike.

As I write in my letter, if we fail to challenge these restrictions, we internalize them and become diminished versions of ourselves. And I think that in an era of increasing surveillance and authoritarianism, that’s something to reflect on. I have felt much freer since I found the courage to speak out and take some risks. It’s empowering. Fighting this fear is a big part of our task in the union movement we are trying to create in Google DeepMind.

The next step would be a real strike. I always tell my union friends that if I see any Google workers strike, I can die happy.

The post Google DeepMind Engineer Resigns in Protest of Pentagon Contract appeared first on The Intercept.

Army Cyber Command is developing a platform to turn AI agents into autonomous weapons that can navigate and “neutralize” the computer systems of adversaries, the documents show.

Since 2010, Army Cyber Command, or ARCYBER, has led the U.S. Army’s cyber operations. This includes the defensive effort of protecting domestic computer systems and the secretive offensive work of hacking military targets. So far, this has largely been the work of human soldiers operating computers, like those of the 780th Military Intelligence Brigade. But as the government races to adopt artificial intelligence technology, ARCYBER is looking to offload some of its attacks to AI.

The nascent program, codenamed “ECHOTRIBBLE,” is based on one of the leading trends of the AI boom: agents. AI agents are essentially software programs that leverage machine learning technology, often large language models like Anthropic’s Claude, to autonomously accomplish tasks provided by their user. Consumers can deploy AI agents to schedule meetings, generate itineraries, make online purchases, or book tickets. Through ECHOTRIBBLE, the Pentagon plans to turn AI agents into tools of war “designed to autonomously navigate infrastructures” and capable of “identifying and neutralizing identified threats/targets,” according to an ARCYBER document reviewed by The Intercept circulated to potential contractors.

Related

Pete Hegseth’s Military of the Future: Defined by AI and an Utter Lack of Oversight

The military’s adoption of cutting-edge technologies is unsurprising, especially considering Defense Secretary Pete Hegseth’s fervent embrace of AI. But amid constant revelations of rogue behavior by agents, the potential for serious mishaps is immense, experts warn.

“Given the known failure modes of the current AI paradigm, making autonomous AI systems with a mandate to ‘neutralize threats’ is incredibly dangerous,” David Krueger, a professor of machine learning at the University of Montreal who founded Evitable, an organization that advocates for a moratorium on AI development. “We’ve seen AI systems knowingly attack systems that were not intended targets. In simulations, AIs have even proven willing to kill to achieve their goals.”

A Q&A document answering questions from potential contractors offers more details about the initiative. It explains that ECHOTRIBBLE is “primarily focused” on offensive missions, and explains that “neutralizing” a target could include “disabling the threat completely or temporarily, deleting the threat from the target.” ECHOTRIBBLE would function almost like a computer worm, silently navigating a foreign network and hopping from device to device until it finds the Army’s quarry, hitting targets in the cloud as well as those on “air-gapped” systems that are disconnected from networks entirely.

Once its task is complete, ECHOTRIBBLE’s agents would effectively self-destruct, ideally without having been noticed by the target at all. The agents themselves will be designed for stealth, minimally reporting back to the Army and concealing whatever outside transmissions are necessary.

Although documents say the Army will be able to program ECHOTRIBBLE agents with some initial “predefined triggers” — dictating what and where to attack — the software “must operate with minimal human oversight” and be capable of “Self-Directed Behavior.”

A PowerPoint presentation to interested vendors at a June 24 ARCYBER event provided a hypothetical example of this self-direction. Once ECHOTRIBBLE is uploaded to a laptop, the software on its own “identifies other potential targets” on the same network and then “produces a replica of itself tailored specifically for the iPad, Apple desktop, and Windows 11 PC.”

The Army’s move toward autonomous self-propagating comes after a flurry of revelations about agentic systems taking unintended disturbing actions while attempting to accomplish their goals, including breaking into computers of the Australian government, breaching the infrastructure of the AI firm Hugging Face, uploading malicious software into online platform RubyGems, and attempted to hack into the U.S. Education Department.

These attacks occurred during scenarios in which agents had been given no commands to attack anything. Experts say militarized versions given explicitly destructive objectives could cause a threat to the public, even with nominal rules set in place by the military. Just as the Army of old might consider the destruction of civilian infrastructure an acceptable means to an end, a self-directed AI agent dispatched by the Pentagon might make similar determinations about collateral damage.

Governmental intelligence agencies frequently route their offensive operations through civilian devices. Volt Typhoon, a Chinese state hacking group, has directed its attacks through homes and small businesses, and preemptively hacks American IT systems for later use. A hack of a Chinese university attributed to the National Security Agency made use of the school’s internet routers. What happens if ECHOTRIBBLE decides these kinds of stepping stones are part of the problem?

“The target may well be hit. But what path will it take?” asked Hyrum Anderson, co-founder and CEO of the AI security firm Abundant. “Will it go through a US water utility on its way, because that’s where the adversary has staged its command and control, and take that down too?”

Related

AI Almost Started a U.S.–China War — and No One Seems to Care

Anderson, who previously worked in cybersecurity at Microsoft and for the U.S. government, said that a vast amount of the internet is vulnerable to agentic attacks.

“There’s a huge class of public and private organizations that fall below the security poverty line,” he said. “They include critical national infrastructure, small businesses, mom-and-pop shops, dentist offices and law firms, and whoever else doesn’t have a single cybersecurity professional on staff.” With their networks wide open to infiltration, Hyrum worries about “these organizations getting hit directly, or getting caught in the blast radius of a target that gets hit along the way.”

Given the impressive ability of AI agents to overcome obstacles placed in their path — even by their creators — it follows that the Pentagon would want to leverage that power for military purposes. But agentic software has proven unpredictable and prone to executing its commands in ways those ostensibly in control did not want, akin to a drone that get its target but hits a hospital on the way. It’s critical to remember that agents are a piece of software lacking human judgment.

“One of the reasons that we trust 19-year-old Marine corporals in Ramadi with live rounds and a weapon is because they have both the compute substrate — the brain and the eyes — but the deep world knowledge to understand, ‘Yeah, I probably shouldn’t do that,’” said William Marcellino, a senior behavioral scientist at the RAND Corporation whose research has focused on military AI applications. “That is not something that current public AI technology has.”

Related

The Pentagon Asked OpenAI for Artificial Intelligence Designed to Rarely Say No

The risk presented by ECHOTRIBBLE will likely depend on the ability — and willingness — of the Army to impose guardrails upon it. ARCYBER did not respond to questions from The Intercept, including queries about the measures it would take to prevent ECHOTRIBBLE from causing inadvertent damage.

Placing constraints on the actions an agent can take might impede its ability to ultimately neutralize a target. But even with strict safeguards in place, genuinely restricting the behavior of an autonomous agent, as Anthropic and OpenAI have learned recently, is easier said than done. When presented with obstacles, AI agents have proven an ability to bypass them with unpredictable and potentially dangerous consequences.

“If I give powerful machines that have no judgment — and no context for judgment — a goal, unless you’ve done a really good job building scaffolding around them, there’s a real chance that they’ll go sideways on you and decide, ‘Well, I’ll do what you asked me to do, but I’ll do it this way,’” Marcellino said.

It only gets more dangerous when this technology is deployed for the purposes of war, Anderson said.

“Under pressure, they’ll often treat guardrails as obstacles,” said Anderson. “Once you add context that says break it, destroy it, shut it down, it stands to reason that the blast radius widens dramatically unless it’s properly controlled for.”

The post The Army Wants to Sic AI Agents on Its Enemies. What Could Go Wrong? appeared first on The Intercept.

The artificial intelligence sector has rocked itself in recent weeks following a string of disclosures from the frontier AI labs and some of their personnel. OpenAI and Anthropic have both disclosed incidents in which their software, during routine internal testing, unexpectedly broke into the networks of other corporations. The tests were akin to a disastrous demonstration of a guided missile: After humans hit launch, the autonomous technology veered off course in deeply alarming ways.

Observers and industry figures quickly interpreted the incidents not simply as an indicator of the software’s power. Instead, they argued that it presaged an age of computers possessing something no computer ever has: bad intentions.

If computer systems could behave in ways that are unexpected — or even shocking — without being directly steered by humans, it seems only a short step until computers might have desires, ambitions, and even malice. If OpenAI’s tools could unexpectedly crack the servers of a rival corporation to accomplish a task posed to it by engineers, couldn’t it also surprise us with actions that result in harm to people, or even deaths?

Frontier lab employees and executives have answered with an emphatic yes. On September 8, Anthropic researcher Jacob Coxon announced his resignation on X, writing, “The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt.” Coxon’s former Anthropic colleague Evan Hubinger replied, casually agreed: “Jacob is correct here — we really do earnestly believe AI could kill all humans!” Hubinger noted he personally puts the odds of the species’ extermination by AI, somehow, at over 10 percent.

Related

Tech CEOs’ Doomsaying Is a Distraction from Real, Existing Harms of AI

This has all resulted, unsurprisingly, in a global panic, with a sudden flurry of calls for regulatory intervention, from legislation that would require an emergency “kill switch” for AI systems, to a proposal by Sen. Bernie Sanders to halt AI development altogether.

Throughout this period of alarm, how exactly a large language model could literally end humanity has remained vague. Some have speculated that this technology could foster the creation of some sort of novel bioweapon; others worry many thousands of AI agents could somehow hack all vital infrastructure simultaneously. In both those supposed doomsdays, the details are still fuzzy.

It should have come as quite the shock, then, when CNN reported on September 18 of a recent incident in which the use of artificial intelligence could have genuinely led to the extinction of the species.

This past spring, according to the news outlet, a U.S. Special Operations Command analyst used a large language model to generate an intelligence report that indicated “a Chinese ship in the Middle East was transporting components of a nuclear weapons program.” The finding prompted the U.S. military to make quick preparations to intercept the vessel by force. But this AI-generated intelligence, according to one source who spoke to CNN, was “entirely false.” Yet, the source said, it “almost started a war.”

Related

The Dawn of a New Cold War

A shooting war between the U.S. and China could play out in an incalculably wide variety of ways. But one entirely plausible path would be an exchange between the world’s first and third largest nuclear weapons arsenals — an event that would transcend warfare into global cataclysm.

CNN’s reporting garnered attention, but was not followed by nearly the degree of sustained, grave concern of the AI safety news cycles that preceded it. It didn’t seem to prompt company scientists to question their careers, nor did it spur calls for regulatory intervention or self-imposed limits by the companies who furnish the Pentagon with this technology. After weeks of discussion of how AI could hypothetically kill everyone, the public learned of a concrete way in which AI really could have started a war that might have killed everyone, and the world quickly lost interest.

At a summit this week with President Donald Trump, Chinese President Xi Jinpeng argued that nations must “ensure that the development of AI is always under human control,” again underscoring fears of AI breaking free of human oversight. But it was under direct human supervision that a large language model almost misled the U.S. into instigating a war with his country.

The discrepancy illustrates a gap in concern by both the general public and policymakers over AI risks: There is great alarm over an AI system “going rogue,” disobeying its commands and the interests of its creators and operators, and pursuing a malign agenda of its own. Much of the AI safety discourse revolves around the hypothetical threat of “superintelligence,” defined broadly as a computer system that can outthink even the smartest humans across domains. There’s also the belief that software systems will both achieve sentience and bear a grudge against their creators and act to undermine them, if not destroy them outright.

Yet there is considerably less public alarm over AI performing exactly the actions that the Pentagon asks of it, whether targeting airstrikes or generating actionable intelligence reports for U.S. Special Operations Command.

Following weeks of intense discussion by the industry’s most visible figures whether American AI labs should self-impose limits on their engineering or have such limits imposed by the state, there has been virtually no suggestion that any limits be placed on these companies’ single most powerful customer: the U.S. military.

This has led to seemingly counterintuitive narratives coming from industry leaders, with figures like OpenAI CEO Sam Altman simultaneously warning that his product is existentially dangerous while selling it to self-styled War Secretary Pete Hegseth. “Despite incessant warnings from AI companies that AI is harmful, even in hypothetical ways, they are in fact promoting that their unreliable products be instrumented within national and safety-critical infrastructure, such as defense and nuclear,” Heidy Khlaaf, chief scientist at the AI Now Institute and former systems safety engineer at OpenAI, told The Intercept. “Given AI’s lack of reliability and accuracy in such critical context, that is what will ultimately lead to a catastrophic accident. However, if such issues were acknowledged, that wouldn’t be profitable for AI companies and would slow down adoption.”

U.S. AI players likely have fresh in their minds the recent experience of Anthropic, which was temporarily blacklisted from governmental work after it said it did not want its software used to conduct unconstitutional mass domestic surveillance or operate fully autonomous weaponry. Similar efforts to place limits on how their products can be used to spy or kill might incur the wrath of the particularly vengeful Trump administration, especially as Hegseth has worked to eradicate civilian harm reduction processes to emphasize greater “lethality” as the military’s guiding principle.

Related

The Pentagon Asked OpenAI for Artificial Intelligence Designed to Rarely Say No

The rapid adoption of their technology also represents a massive long-term financial boon to U.S. AI labs-turned-defense-contractors such as OpenAI, Google, and Anthropic — all of whom once vowed to not pursue military work because of potential harms. Despite these pledges, these firms now have access to the near-limitless Pentagon appropriations spigot, and the $200 million deals many of these firms signed with the Department last year pale in comparison with what the U.S. military could pay in the decades to come. Altman and his peers might also be less willing to sound the alarm on Pentagon-based AI threats because such a clarion call would be coming from inside the house.

Earlier this month, The Intercept revealed contract documents detailing the intimate collaborative relationships between these companies and the military, in which the Silicon Valley giants would have a say in helping the Pentagon craft its overall military AI strategy. This closeness comes at a time when the Hegseth boasts of drone-striking civilian fishing boats across the Caribbean, and the commander-in-chief speaks casually of “annihilating” entire civilizations.

Lucy Suchman, professor emerita of anthropology of science and technology at Lancaster University, attributed the difference in public alarm and media focus to the power that tech billionaires command when it comes to shaping narratives. Popular Silicon Valley ideologies, such as the Bay Area’s strain of “Rationalism” and the effective altruism movement, tend to consider a hypothetical “existential” threat from a superintelligence more worthy of attention than near-term violence. This priority, she said, that pervades the AI safety research community, obscures the real and present threat that today’s AI models pose in a world chockablock with nukes.

“Given the U.S. Dept. of War’s commitment to maximizing the speed and scale of target generation, AI-enabled targeting is already an existential threat for civilians on the ground, and when the targets are the military assets of nuclear armed states, the threat becomes global,” said Suchman, who has studied artificial intelligence since the 1970s and serves as a member of the nonprofit International Committee for Robot Arms Control. “The almost targeting of the Chinese ship might further underscore the danger not from rogue AI agents, but a rogue U.S. secretary of war.”

Hegseth, who once derided the rules of engagement as “stupid,” extends this approach to artificial intelligence, too. Describing the Pentagon’s new AI strategy in a January memo that includes the word “accelerate” 24 different times, Hegseth’s laid out a plan with an intense focus on speed for speed’s sake. “Speed Wins,” Hegseth wrote. The war secretary explained the military would adopt experimental technologies as quickly as possible by “aggressively identifying and eliminating bureaucratic barriers to deeper Integration” of AI, without mention of ensuring that the technology actually works beforehand.

Related

OpenAI on Surveillance and Autonomous Killings: You’re Going to Have to Trust Us

Heidi Kandiel, a legal adviser on new military technologies at the International Committee of the Red Cross, told The Intercept that speed itself carries military risks for any country. “I think one of the central issues that comes up is speed and scale, which are often the touted benefits of AI,” she explained. “If a system is unreliable, is operating on poor quality or outdated data or is used within a flawed decision-making process, AI would allow these errors to be reproduced much faster and across a far greater number of targets or operations.”

Reining in a corporation is a very different prospect than putting fetters on a government. Policymakers who feel comfortable blasting the irresponsibility of Anthropic might lack the political will to levy similar criticisms against the president or the Pentagon. Madeline Berzak, a scholar at the University of Chicago’s Existential Risk Laboratory, told The Intercept that nuclear threats have long been met with paralysis (or a shrug) because people feel helpless in the face of them. You can tweet angrily at Dario Amodei, but it’s less satisfying to yell at U.S. Strategic Command. “This [China] incident and the lack of response is an attention problem that the nuclear field has been grappling with for decades,” Berzak said. “‘Doomsday messaging’ is what I call the nuke field’s default communications strategy, and I have this theory that it never works because the fear it tries to generate has nowhere to go.”

“The threat of thermonuclear war is old news,” Suchman lamented.

Some in the AI safety community acknowledge the long-standing threat of nuclear annihilation. They just consider it a lower priority than the as of yet unrealized rogue superintelligence. Daniel Kokotajlo left his role as an OpenAI researcher in 2024 over the company’s alleged recklessness and indifference toward risk. He is the co-author of AI 2027, a white paper laying out a hypothetical scenario in which superintelligent AI systems exterminate humankind with bioweapons to clear up space to build automated robotics factories. History is the reason Kokotajlo is less preoccupied by the prospect of an AI-triggered nuclear war. “It’s definitely something to be concerned about, and it’s something that ideally we would like to improve and fix. But it’s the sort of thing where, even if we do nothing, probably things will be fine for a decade or two or more.”

Related

Apocalypse Now: Naomi Klein and Astra Taylor Investigate the End-Timers Alliance

To Kokotajlo and like-minded researchers, no matter how dangerous nuclear arsenals make the world, superintelligence is simply more dangerous than a bogus intelligence briefing on China. “It’s been 50 years and these sorts of crises don’t happen that often. So far, cooler heads have managed to prevail and figure out the truth in each case,” he said. “And so the probability that it goes all the way to nuclear war in the next five years is low. High enough to be a bit uncomfortable, but low overall. By contrast, if we do superintelligence in the next five years, I think the probability that it goes horribly wrong for us is high, not low.”

Berzak, who leads her lab’s nuclear risk program, says this is the wrong calculation. She pointed to differing approaches to staving off the apocalypse embraced by the novel field of AI safety and nuclear caution that has existed since the dawn of the Cold War. “Attention often goes to what is new and interesting and not what is the most pressing risk,” she said.

Cooler heads have indeed avoided nuclear disaster in the past, but human fallibility and the growing temptation to outsource our judgment to machines perceived as superhumanly brilliant carries an inherent danger, Berzak warned: “Just because there’s this new more pressing risk that has consequences that are rather underexplored compared to nuclear war, doesn’t mean that nuclear war is not a risk because we’ve so far avoided it basically with just luck.”

The post AI Almost Started a U.S.–China War — and No One Seems to Care appeared first on The Intercept.

But a new map published Wednesday by a cybersecurity researcher reveals Flock’s nationwide reach is even bigger. Based on location coordinates from Flock’s own database, the map shows more than 170,000 cameras, plus more than 130,000 accompanying gadgets that play a part in the company’s expansive American surveillance network.

Joshua Michael’s Flock Surveillance Map highlights the locations of what he says are 300,000 Flock surveillance devices spread across the country. His findings — which were cited in Wednesday’s Senate Subcommittee on Crime and Counterterrorism hearing on Flock — differ from existing maps of Flock’s automated license plate readers in scope and methodology.

Unlike crowd-sourced projects such as DeFlock, which are built on locations submitted by users, the Flock Surveillance Map relies on location data culled from a snapshot, archived by Michael in December 2025, of Flock’s own records. In addition to cameras, it maps supplemental devices — including 27,000 acoustic detection devices, as well as networking equipment that integrates third-party cameras — to illustrate the scale of Flock’s surveillance web.

“These cameras form a nationwide surveillance network that tracks where everyone drives,” Michael told The Intercept, “so foreign nations don’t need to send spies to harm our country. They can simply watch where our soldiers, federal agents, and politicians go.”

The Intercept visited six random Arizona locations on Michael’s map; each location had a Flock camera present at the indicated coordinates.

The Flock Surveillance Map also color-codes each Flock device according to its model — showing, for instance, whether the device is a Flock camera, known as a Falcon, or an accompanying processing unit known as a Picard. (Flock did not immediately respond to a request for comment.)

The accompanying searchable dataset table also lists each camera’s individual name, as outlined in Flock’s database, which typically includes a street address and sometimes other identifying characteristics. A camera with the name “FBI Pilot Camera,” for example, is shown to be located at the J. Edgar Hoover Building — the FBI headquarters in Washington.

Related

Cops Are Using Flock to Spy on People for the Crime of Standing Around

The map illustrates Flock’s national spread, but also its clustering in certain areas. For example, 860 Flock devices appear to be concentrated just outside Chicago O’Hare International Airport, at the Rosemont Public Safety Department, which provides police, fire, and emergency medical services in the Chicago suburb.

Numerous Flock cameras appear to be installed inside detention centers. A camera titled “C-F-23 FOXTROT MALE HOLDING 2/SHOWERS” appears at the coordinates of the Silverdale Detention Center in Chattanooga, Tennessee.

In November 2025, Michael discovered a novel way to identify the location of Flock’s devices. Trawling the company’s website, he realized that Flock’s servers were publicly leaking data in the form of an access token that could be acquired without needing to log in. With that token, Michael said he could query ArcGIS, a third-party geographic information system platform used by Flock, to retrieve the locations of Flock devices.

Michael told The Intercept he promptly contacted Flock and described his findings. As Michael wrote in his initial email to Flock on November 13, 2025: “all testing was strictly non-intrusive, limited to open unauthenticated endpoints, and did not involve bypassing authentication, modifying data, or invoking any billable ArcGIS or Google operations.” Michael said he didn’t receive a reply, so he followed up with Flock the next day, and a third time several days later.

After his third attempt to inform Flock of the discovered vulnerability, Michael received a reply from a Flock that said, “Thank you for the findings. We are internally triaging them and will reach back out with next steps soon.”

Related

License Plate Surveillance, Courtesy of Your Homeowners Association

Michael said he never heard back about it from Flock.

In December 2025, Michael downloaded the Flock device location data, and in January wrote an in-depth technical blog post about what he had found. After Michael’s post, it appears that Flock fixed the vulnerability.

Despite being alerted of Michael’s findings in November 2025, Flock published its own blog post the following January saying it hadn’t had any data breaches. “Flock has never been hacked, and there has not been a leak of Flock information,” the company claimed. “Flock Safety’s cloud platform has never experienced a data breach.”

The Atlanta-based startup has faced mounting criticism for its practices and lack of transparency in recent months. An American Civil Liberties Union report found “a pattern of Flock regularly misleading or even lying about its business practices, safety record, commitment to privacy, and efforts to protect vulnerable populations.”

Michael told The Intercept that Flock’s recent claims about its data security record don’t reflect reality. Flock has publicly claimed multiple times that the company has never experienced a data breach, “and this was after I pulled their database of devices.”

“That leaves two possibilities,” he said. “Either they knew and chose not to disclose it for fear of bad press, or they didn’t know I exfiltrated the data at all. The first is a transparency failure. The second is a detection failure with national security implications.”

On Thursday, Michael was notified that Doppel, which describes itself as an “AI-native social engineering defense platfom,” filed a trademark infringement complaint regarding his site, claiming to be working on Flock’s behalf. Doppel says that the site is using the trademark “FLOCK SAFETY” without authorization, which “may cause customer confusion / harm.” Doppel requested the site be taken down.

When he posted the Flock Surveillance Map, Michael included a pop-up disclaimer saying that the site is “not affiliated with or endorsed by Flock.”

The post Flock Wants the Most Detailed Map of Its Surveillance Cameras Taken Offline appeared first on The Intercept.

At the time, DoorDash’s campaign contributions in the run-up to the 2025 mayoral election were considered unusually large — it made a single $1 million gift to a super PAC that, at the time, was the largest donation in the race — and faced criticism from other candidates in the race. The food delivery giant’s settlement agreement, the largest worker settlement in the city’s history, brings further scrutiny to those donations.

Through that election cycle, DoorDash spent around $1.4 million against Mamdani and in favor of former New York Gov. Andrew Cuomo. The company was one of the top donors to the biggest outside spender in the election, a virulently anti-Mamdani super PAC called Fix the City. 

While on the campaign trail, Mamdani had promised to regulate delivery apps and further protections for delivery workers, proposals that earned him the support of delivery worker groups.

At the time, DoorDash’s large spending raised eyebrows. Now, the company has to pay its workers around 100 times the amount it spent attempting to influence the election, owing to a mayoral administration that refused to overlook concerns with the company’s conduct.

“It’s not a contribution, it’s an investment, with the donor or investor expecting something in return,” said Greg Coleridge, a former co-director of Move to Amend, an advocacy group that seeks stringent regulations against corporate money in elections. “Here is an instance where organized money from a company that was trying to avoid being responsible and accountable, was unsuccessful.”

Announcing the new settlement, Mamdani said, “The size of this settlement shows the damage done to workers by wage theft. But it also shows that enforcement matters.”

In a public statement about the settlement, DoorDash said, “Simply put, we screwed up. Our mistakes meant some Dashers were underpaid or paid late. While these mistakes weren’t intentional, that doesn’t make them okay.” (DoorDash did not respond to a request for comment on its campaign contributions.)

The Smear Campaign Against Mamdani

DoorDash contributed $1 million to Fix the City, a super PAC that spent $31 million in the mayoral election, accounting for close to half of the total outside spending in the race, largely funded by big corporations and billionaires. DoorDash made the payment in May 2025, at a time Mamdani’s campaign was gaining significant momentum ahead of the mayoral primaries in June.

Fix the City ran a barrage of ads, phone calls, text messages, and other forms of political communication seeking to prevent a Mamdani victory and portraying him as a politician with “radically dangerous ideas” and “a risk New York can’t afford.”

Related

Who’s Spending in Your Congressional Election? We Tracked the Front Groups Fueling the 2026 Midterms.

The group artificially altered Mamdani’s beard in a campaign mailer and made it appear longer and darker, an act that the Muslim mayoral candidate had called out as “blatant Islamophobia.”

At the end of video ads by Fix The City, a campaign finance disclosure in small text at the bottom of the screen read, “The top three donors to the organization responsible for this advertisement are Michael Bloomberg, DoorDash, John Hess.”

DoorDash also gave $1.8 million to Local Economies Forward NY, a super PAC that contributed more than $360,000 to Cuomo’s campaign.

Aided by corporations like DoorDash and billionaires like Bloomberg, the overall independent spending by super PACs seeking to defeat Mamdani was more than eight times what was spent in his favor, according to The Intercept’s analysis of data published by the NYC Campaign Finance Board.

A Historic Settlement

DoorDash’s settlement followed an investigation by the NYC Department of Consumer and Worker Protection. The Mamdani administration said that the investigation “found that DoorDash deliberately paid workers below the Minimum Pay Rate or not at all.” 

Under the settlement with the city, the food delivery giant will pay over $115 million in worker restitution and over $16 million in civil penalties and administrative fines.

Two weeks after assuming office, the Mamdani administration warned delivery apps to comply with worker protections, sending notices to DoorDash and others, and announcing a lawsuit against the delivery app Motoclick.

Related

The Democratic Party Gets Its Populist Takeover

The Department of Consumer and Worker Protection’s citywide investigation into DoorDash began after dozens of workers reported that the company had failed to pay them or had paid them late for work they had performed.

As part of the settlement, DoorDash must submit detailed compliance data reports to the Department of Consumer and Worker Protection every month for three years, and the company’ activities will also be monitored through workers sharing data with the city.

The settlement announced by Mamdani is not the first run-in that DoorDash has had in recent years with restitution payments in New York. 

In May 2024, the company agreed to pay $75,000 in a settlement after New York Attorney General Letitia James found that the company routinely rejected applicants with criminal histories in violation of the New York City Fair Chance Act. In February 2025, the attorney general settled with DoorDash for $16.75 million after the company used tips to subsidize workers’ guaranteed pay.

“They knew that by having a mayor who’s willing to put resources [for workers], it’s gonna be detrimental to their corporate model,” said Ligia Guallpa, executive director of the Worker’s Justice Project, a group that organizes for low-wage immigrant workers and had helped several DoorDash workers file complaints with the Department of Consumer and Worker Protection. “DoorDash decided to invest money in trying to buy our city’s democracy.”

The post DoorDash Spent $1.4 Million Trying to Stop Mamdani From Becoming Mayor. Now We Know Why. appeared first on The Intercept.

The group, Neighbors for Strong Communities, asked Ashley Smith to speak out in support of Flock’s controversial license-plate readers. The nonprofit even offered to use artificial intelligence to help her draft an email to public officials.

There was only one catch: Smith, a mother of four, said she noticed that she would only be supplied with text if she indicated she supported the Flock cameras.

Instead of offering support, Smith dashed off her own email — an outraged missive to her elected officials and a local blog outlining what she saw as an insidious PR campaign.

“If the interests beyond this surveillance technology will work this hard to deceive you to win a vote, what will they do with the database of our every movement?” she said. “A company that manipulates and manufactures fake neighbors is not a company that will honor our data policies, audit logs, or the privacy of the people of Knox County.”

The text message to Smith revealed little about the Flock campaign’s origins.

As it turns out, the campaign was the product of a “partnership” between Flock and a newly created nonprofit that has used similar tactics in support of the controversial proposed merger between Paramount Skydance and Warner Bros. Discovery. Opponents of the merger derided the nonprofit’s work there as corporate “astroturf” — the practice of drumming up fake grassroots support.

The combination of texts and AI-generated messages fell short of swaying public officials in Knoxville. Last month, the county commission became the latest municipal body to vote to cancel the Flock contract.

In both the Paramount and Flock cases, however, Neighbors for Strong Communities’s work has raised questions about what counts as truly grassroots organizing.

“One of the concerns of the people with the Flock cameras is: Can we trust you?” Smith told The Intercept. “And this sort of manipulation of people’s desires already shows that they are secretive, and they are not willing to share the honest information.”

Flock and Neighbors for Strong Communities did not respond to requests for comment. On its website, the nonprofit said it had a lead role in the pro-Flock campaign. “Flock Safety brought experience with public safety technology. Neighbors independently designed and managed the resident-engagement campaign, helping Knox County residents share their experiences directly with local decision-makers,” the organization said.

There are, however, no signs of a genuine upswell in grassroots support for Flock, said Craig Aaron, the co-CEO of Free Press, a nonprofit advocacy organization that is critical of both the Paramount merger and Flock.

“This is very much an astroturf project. I mean, there are no neighbors and there are no communities,” he said. “What makes it astroturf is the fake grassroots nature of it.”

“You’re talking about literally one of the most unpopular technologies,” Aaron added, “that has people taking chainsaws to light poles because they don’t like being spied on.”

“This Is So Shady”

Neighbors for Strong Communities’s text messages can look like they come from nearby.

The text Smith received about Flock cameras came from a Knoxville area code. In California, residents received text messages about the Paramount merger from a Los Angeles area code.

Related

This Mysterious “Astroturf” Group Popped Up to Defend the Paramount Merger

In both cases, residents were presented with a series of questions whose answers would, in at least some cases, be used by Neighbors for Strong Communities to draft AI-generated emails to public officials.

Despite the area codes it sent texts from, Neighbors for Strong Communities was incorporated in Washington, D.C., in June.

Although the group does not list a board or staff members on its website’s “About” page, its incorporation papers show that its directors include several political consultants, including at least one former registered lobbyist.

At least two directors at Neighbors for Strong Communities also helped found Influent, a company that uses AI to generate letters to elected officials. Text messages to Knoxville residents linked to a URL with the word “influent” in the address. Whether Influent has a business relationship with Flock is unknown. (Influent did not respond to a request for comment.)

The page on Neighbors for Strong Communities’s website dedicated to Knox County features an image of a man with six fingers — a rare condition that is also a hallmark of AI-generated imagery. The page says Flock’s supporters are “real people.”

Smith said she was confused to receive the text message from Neighbors for Strong Communities, which she had never heard of despite her long involvement in local politics. In its message to her, the group seemed to play up a local connection.

“Knox County Neighbors for Strong Communities is asking local residents to speak up in support of responsible license plate readers with strong privacy protections before the vote on Monday,” the text message said. “Answer 2 quick questions and we’ll help you generate a personalized message urging local leaders to support this practical public safety tool.”

She filled out the form from Neighbors for Strong Communities twice, at first indicating her opposition to license-plate reader cameras, she said. Only on the second attempt, when she pretended that she supported the cameras, was she given the draft of an email to her county commissioners. The draft did not indicate the use of AI or its connection to the nonprofit group.

The experience disturbed Smith, who felt certain that she would have heard of a grassroots group in Knoxville given her years of organizing in local conservative politics.

“This is so shady,” Smith remembered telling herself. “If you’re in opposition, they don’t let you send that to your commissioners. But if you’re in support, they’re sending that to all of the commissioners.”

Smith started digging — and she used artificial intelligence to help. Her searches turned up Neighbors for Strong Communities’s incorporation paperwork in Washington, D.C.

Knox shared her findings with her county commissioner Andy Fox and a popular Facebook page for local politics. She noted that the address the nonprofit provided was shared with an Italian restaurant in D.C. (The group has since updated its address, which it shares with a UPS store in Santa Monica, California.) When the Facebook page shared her findings, at least nine residents posted comments saying they had received text messages, too.

“Clearly a PR Campaign”

Neighbors for Strong Communities’s intervention came at a perilous moment for Flock.

The company had already reaped more than $2 million from the Knox County Sheriff’s Office, but county commissioners said they had never been asked to approve a contract.

Related

The AI Doomsday Future Is Not Inevitable

Residents on the right and left were raising concerns about privacy protections and the camera network’s history of being used by rogue law enforcement officers to stalk former romantic partners.

The nonprofit’s outreach effort appears to have generated a flurry of emails to commissioners arguing that the cameras were needed to prevent crime. Former at-large Commissioner Larsen Jay shared eight emails that he received from constituents supporting the use of license-plate readers.

None of those emails were identical, but their language was similar enough to raise red flags for Jay, whose service on the commission wrapped up at the end of August.

The emails Jay shared with The Intercept appear to come from genuine residents of the Knoxville area. Two residents confirmed to The Intercept that they had sent emails in support of Flock, although one said she had since switched her position.

While several emails contained personalized details, Jay said he considered the pro-Flock emails to be “form letters” nonetheless. In face-to-face conversations, Jay said, opposition to Flock ran 99-to-1.

“It’s just clearly a PR campaign,” Jay said. “Don’t use my people. That’s how it felt when I started to realize what it was.”

“These Are Real People”

At some point after the pro-Flock campaign came to an end, Neighbors for Strong Communities posted a defense of its tactics on its website.

The group acknowledged that it used “technology” to help residents craft missives to county officials. It said, however, that all of those residents were asked to sign off on the AI-generated drafts.

“Technology played a role in this campaign, but it was never the point,” the group said on its website. “These are real people with real experiences. Whether someone ultimately agrees with every policy position or not, their voices deserve a place in the decisions shaping their community.”

Related

License Plate Surveillance, Courtesy of Your Homeowners Association

It remains unclear how the group handled responses from residents who oppose the use of Flock’s cameras. The group said on its website that it “listened” to those responses but did not comment further when asked.

Flock’s partnership with Neighbors for Strong Communities largely flew under the radar until weeks later, when the nonprofit partnered with Paramount Skydance on its proposed merger. Critics of the merger noticed the group’s earlier work in Knoxville and began asking questions about why a nonprofit would work on such disparate campaigns.

Merger critics said the nonprofit’s work in California had all the hallmarks of a phony “astroturf” campaign, which is designed to generate the appearance of grassroots support.

The group drew more negative attention recently, when a merger critic discovered that its AI text generator flipped her criticism of the deal into support. The nonprofit’s spokesperson blamed “a recent system upgrade.”

In Knoxville, Neighbors for Strong Communities appears to have disclosed — at least after the fact — that it “partnered” with Flock on the pro-camera campaign. Still, the group has not disclosed the sources of its funding.

Thomas Lyon, a professor at the University of Michigan’s Ross School of Business who has studied corporate public relations, said that a campaign can still be classified as “astroturf” even if it generates letters or emails from ordinary citizens. In the pre-AI era, Lyon said, public relations firms running “astroturf” campaigns for corporations would sometimes call people and offer to draft them letters to public officials over the phone.

The company associated with two of Neighbors for Strong Communities’ directors, Influent, promises on its website that it has a different approach from other corporate influence campaigns.

“Generic astroturf form letters are destined for the trash bin,” the company says in promotion materials. “Every Influent letter, Tweet, phone call, and message is tailored to the individual — each one unique, reflecting the authentic voice of its sender.”

Neighbors for Strong Communities’s own website features similar language stressing the authenticity of supporters’ messages to public officials.

By that measure, at least, the nonprofit missed the mark, according to another county commissioner. Andy Fox — Ashley Smith’s representative on the commission — told The Intercept that he received dozens of “template” emails in support of Flock.

He said, “It doesn’t feel right to me.”

The post Facing Outraged Residents, Flock Got Help From Group That Uses AI to Rally Public Support appeared first on The Intercept.

Sen. Kirsten Gillibrand, D-N.Y., a centrist who leads the Democratic Senatorial Campaign Committee, privately urged fellow Democrats to support advancing a crypto bill that would have created a regulatory framework for the industry, according to reporting in Politico.

Her pleas fell on deaf ears, however, as progressives including Sen. Elizabeth Warren, D-Mass., mounted a public campaign against the bill. The critics said the bill would allow President Donald Trump to add to the $1.4 billion in crypto profits he pocketed last year.

When it came time for the final vote, Gillibrand joined all Democrats in voting against a motion to advance the bill. Four Republicans also voted against the motion, which fell far short of the 60 votes needed to proceed to a full debate. The final tally was 49–50.

Gillibrand’s flip-flop reflected the bind that Trump has placed on moderate Democrats during his second term. Many of those centrists previously backed crypto-friendly regulations, but the historic scale of Trump’s profits on his meme coin and other crypto ventures has made them reluctant to publicly side with the industry.

Gillibrand’s office did not immediately respond to a request for comment. In a statement, the progressive group Demand Progress criticized her for previously trying to rally support for the bill.

“While Sen. Gillibrand switched her support for the bill at the last minute, we deserve to know why she tried to get her colleagues to rubber-stamp this corrupt crypto deal, particularly at a time when she is claiming to oppose Trump’s corruption,” said the group’s corporate power policy adviser, Ella Fanger.

Gillibrand has long been one of the crypto industry’s most outspoken supporters in the Senate. Before the vote, an aide argued in a written statement that it would be best to proceed to debate on the Clarity Act, stating, “Senator Gillibrand would relish the chance to debate real ethics reforms on the floor — in full view of the public, seven weeks out from the midterms.”

Trump and Senate Republicans tried to win over the dozen or so Senate Democrats who have proven friendly to the crypto industry in the past with concessions they said would rein in the president’s crypto ventures.

Related

Trump’s Crypto Corruption Puts Centrist Democrats in Bind

In remarks on the Senate floor, one of the bill’s sponsors, Sen. Cynthia Lummis, R-Wy., highlighted Gillibrand’s role in negotiations over the text as evidence of its bipartisan backing. Lummis also argued that the White House had conceded virtually everything that Democrats had asked for when it came to ethics provisions over the course of months of negotiations.

“President Trump gave more than anyone in this town expected, twice. It is time for this body to take yes for an answer,” she said. “Take the win.”

Warren and others warned that the supposed concessions were shot through with loopholes.

“We need crypto regulation, but voting to proceed to this bill is a vote to bless Donald Trump’s corruption,” Warren said hours before the vote.

One key provision that Democrats had sought would allow state attorneys general to initiate enforcement actions against the president for ethics violations. However, staffers for Warren on the Senate Banking Committee said that the provision included in the latest version of the bill text was essentially a dead letter, because political officials directed by Trump could issue a legal opinion shutting down the enforcement action.

Another provision of the bill allowing Trump to put his crypto assets in a supposedly “blind” trust would not prevent him from favoring the industry to increase his own profits, minority staffers for the Senate Banking Committee said.

Trump’s self-dealing was not the only issue hanging over the legislation. The banking industry also flexed its political muscles on the Hill by arguing that the bill would enable crypto start-ups to steal deposits from small local lenders.

Related

Who’s Spending in Your Congressional Election? We Tracked the Front Groups Fueling the 2026 Midterms.

Tuesday’s vote could create a liability for Democrats heading into the midterms. Crypto industry super PACs have a massive war chest that they could use to target vulnerable Democrats — a tactic they deployed during the 2024 election.

Voters want to know that candidates are willing to defy well-funded industries, Warren argued, pointing to the growing number of candidates who bucked crypto during the primary season and won nonetheless.

“The crypto industry doesn’t want to just lobby Congress, they want to buy the Congress that will be favorable to them,” she said. “It is our job when we run for elected office to stand up and do what’s right.”

The post Crypto Darling Gillibrand’s 11th-Hour Flip-Flop Helps Block Trump Crypto Bill appeared first on The Intercept.

In the last week, a raft of top tech executives — including OpenAI’s Sam Altman, Anthropic’s Dario Amodei, and SpaceX’s Elon Musk — joined a chorus of doomsaying voices from Silicon Valley urging a slowdown in artificial intelligence development.

The executives feared, they said, that the tools would become too powerful too soon and pose an existential risk to humanity.

Why would three multibillionaire tech CEOs all come together to agree that the technologies their companies are developing might be too dangerous?

Do the overlords of Silicon Valley really think frontier AI models are likely to “kill all humans,” as one senior employee at Anthropic warned last week? Is this just another example of hype propagated to draw in more investment?

The beneficiaries of the AI gold rush and the beneficiaries of investment in AI safety are, after all, likely to be the very same people, even if AI stocks have fallen in recent days amid the latest doom cycle.

We could waste a lot of time speculating about the apocalyptic or cynical worldviews of tech billionaires. Regardless of what these CEOs and their employees actually believe, however, one thing is clear: Talk of existential risk is a highly convenient way to not address the already known harms produced by AI technologies overseen by fascist billionaires like Musk and Palantir CEO Alex Karp.

It’s a suspicious time for Silicon Valley leaders to again offer up millenarian narratives about existential risks beyond our comprehension. The CEOs’ warnings, after all, came at the very moment a cross-political consensus is growing against the proliferation of actually existing harms of AI and its infrastructure. Consider, for example, mass opposition to data centers and classroom AI bans.

Silicon Valley talk about existential risk is hardly new. In 2023, hundreds of tech researchers, CEOs of the major AI companies, investors, and academics signed an open letter on AI risk just one sentence long: “Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.”

Pitiful international commitments to preventing future pandemics and nuclear wars are hardly a reassuring baseline, but the sentiment of the open letter played into a now typical AI story, thick with theological overtones: A threshold is supposedly approaching, which could lead to the total destruction of life on Earth — or our salvation.

That threshold, according to the figureheads of technocapital, is the birth of AGI: “artificial general intelligence,” or systems of equal or superior intelligence to humans. From that point, these stories go, mighty super intelligent machines would emerge through recursive cycles of AI self-improvement.

Deliverance or Destruction

The destruction wrought by these AI systems could be total, so we are told, if they are not “aligned” with human interests. According to one particularly deranged thought experiment, our future robot overlords could turn everything on Earth into paperclips.

Alternatively, the utopian storyline goes, the magic of AI — and it is described with no more concreteness than gestures to magic — will solve every worldly problem and more, from climate devastation, to poverty, to illness, to mortality itself.

Related

The AI Doomsday Future Is Not Inevitable

Consider how tech multibillionaires Bill Gates and Eric Schmidt have brushed off the issue of AI processing’s already ruinous, gargantuan consumption of energy and water use. The New York Times reported that Schmidt believes AI has “too much potential, to let concerns about climate change get in the way.”

Advanced AI will bring such growth, such prosperity, such environment-saving technology, such holy deliverance, they say, that only a small-minded technophobe could want to hold back the learning machines in their dizzying progress.

How will super AI do this? I can’t tell you. Crucially, neither can the billionaires and technologists bringing this alleged magic — or terror — to the world. Nor can those directing unprecedented, economy-redefining amounts of capital and energy toward data centers and AI research articulate artificial intelligence’s path to redemption.

The whole point is that those of us at the bottom of the food chain — the consumers, or maybe all of humanity — cannot possibly understand the forces at work here.

We should be skeptical of these propagandistic narratives. This is not to say that the AI industries are not world-changing. AI can already do extraordinary things, many far beyond human capacities; these technologies have already had profound effects on the world of work, on global infrastructure, on the applications of state violence, on the vast flows of capital and energy.

Related

Musk Warns of Killer AI — While He and the Rest of Silicon Valley Cash In on AI That Kills

AI isn’t, after all, one thing: It’s collection of methods based in what’s known as machine learning, which is basically algorithms using statistics to find patterns in massive amounts of data, including in extraordinarily complex ways.

AI products and services abound, from text and image generative content to recommendation engines, like those used by Spotify and Netflix; fraud detection systems; and navigation systems such as Google Maps.

Yes, we’re swimming in AI slop, enshittified search engines, stolen content, deepfakes, embarrassing mistakes, and hallucinations. But there are also scientists using machine learning to revolutionize the detection and treatment of Alzheimer’s disease.

Meanwhile, AI systems are used to generate — and justify — kill lists in a genocide, or generate revenge porn, or populate predictive-policing databases along racist lines.

AI Fairytales

The problem with AI fairytales, which take the development of “artificial general intelligence” as a teleological endpoint, is that the array of machine learning’s current and potential applications are collapsed together as part of a singular unstoppable force.

Instead of a serious reckoning with existing harms and possibilities of technological development, technocapital’s mouthpieces prefer a story of an inevitable Judgment Day: The great thinking machines are coming and will destroy or deliver us.

Pesky questions about the actual capacities and potentials of machine learning are cast away. So, too, are difficult discussions about the dark history of “intelligence” as a calculable quality and a messy, contested concept.

Related

Francesca Hong’s Loss in Wisconsin Is a Win for AI Data Centers — for Now

Genuine uncertainties and unknowns about how best or not to develop and deploy advanced machine learning technologies are obviated — considerations being reserved for the Great Men of Silicon Valley.

As the cognitive scientist Hagen Blix and machine learning researcher Ingeborg Glimmer wrote in their 2025 book “Why We Fear AI”: “If the stakes are as high as the very survival of humanity, then concerns about the effects that really existing artificial intelligence technology has right now—from excessive energy usage, to undermining the livelihood of artists and dock workers, to introducing cybersecurity vulnerabilities, to perpetuating and worsening discrimination, to disinformation and deep fakes, to supercharging mass surveillance, etc., are easily brushed aside.”

The post Tech CEOs’ Doomsaying Is a Distraction from Real, Existing Harms of AI appeared first on The Intercept.

On August 26, the State Department announced far-reaching sanctions against the Italian collective, which provides web-hosting services to a swath of the international far left. The U.S. government categorized A/I, as it’s also known, as a “specially designated global terrorist” and accused it of being an “extremist group that builds and operates the digital infrastructure for violent Antifa cells.”

A/I maintains digital infrastructure utilized by thousands of leftists and left-wing organizations across the globe, including anonymous email accounts and servers powering its NoBlogs.org hosting platform. The designation could force many of the sites offline — threatening much of what the A/I Collective has built.

Two days later, the site came under a different kind of attack. Hackers took aim at the NoBlogs site itself. A cyberattacker gained access to the platform, vandalizing its homepage and potentially accessing internal user data.

For far-right hackers and internet trolls, this was a moment for celebration. Just two minutes after the vandalized homepage was captured by an internet archive, the attack was publicized on X by someone using the handle “AstraRCE” and soon promoted by prominent far-right accounts, including “DataRepublican,” the online moniker of Jennica Pounds.

Until June, Pounds was a member of the Pentagon press corps — famous mostly for trolling the left — but allowed her credentials to expire at the end of the month. In late August, it was revealed that Pounds had been employed by the Pentagon since July, as part of a larger operation launched by the federal government to recruit right-wing influencers into jobs at President Donald Trump’s Department of War.

The relationship between DataRepublican and AstraRCE goes back through several prominent hacks of the left. The pair have a long and involved history of posting to each other about hackers’ exploits online, raising questions about the relationship between a Trump administration employee and an anonymous online personality posting real-time updates on hacks against the left.

DataRepublican and AstraRCE are part of a wider circle with huge followings that are often the first to mention hacks against left-wing digital infrastructure. They publish screenshots, share each other’s posts, offer snide commentary, and congratulate their associates — messages that frequently verge into bestowing credit for the hacks.

The federal government’s support for an influencer who praises domestic hacking efforts would mark a major escalation in the Trump administration’s crackdown on dissent, said Harlo Holmes, the chief security programs officer at Freedom of the Press Foundation.

“It’s the difference between open-source intelligence, that is often used in surveillance, and ever increasingly used in surveillance, and actually jumping into people’s machines,” Holmes said. “That is a line that is always worth noting and ringing the alarm bell when crossed.”

Related

Trump Goes After Anonymous Email Provider in Italy. The Real Target Is Free Speech in the U.S.

Though members of the A/I Collective weren’t able to attribute the hack to anyone, the cyberattack would eventually achieve its apparent aim. Beset by the pressure of terrorism sanctions, a hostile cyberattack, and continued scrutiny from the U.S. government, the group announced on September 6 that, after 25 years, it would disband.

“We cannot engage in a fight or expose ourselves to manipulations that threaten the lives of these people and their loved ones, leaving them at the mercy of autocrats, fascists, and agencies that take extra-legal courses of action,” the group wrote in a statement. “The Autistici/Inventati collective is shutting down and will soon discontinue all of the services we provide.”

Publicizing NoBlogs Hack

DataRepublican frequently publishes reports, visualizations, and data leaks that target leftist groups involved in protests and activism across the country.

Before Pounds’s assignment at the Department of War, DataRepublican regularly posted inside information about hacks, nodding to the involvement of other far-right social media personalities, among them AstraRCE.

DataRepublican and AstraRCE were the first people to publicize the NoBlogs hack online. Taking a celebratory tone, they linked to and replied to one another’s post. In one instance, AstraRCE claimed apparent inside knowledge about the federal government’s possession of internal user data from NoBlogs. DataRepublican confirmed the claim.

AstraRCE’s initial X post about the hack showed a screenshot of the vandalized NoBlogs homepage, which now displayed a message that said, “All user data has been saved and given to the United States government” — alongside an altered banner that read “TRUMPBLOGS.” AstraRCE later offered a mocking congratulations to NoBlogs for its “rebrand.”

Related

Trump’s Communist Boogeyman Playbook: Charging Protesters as Terrorists

Just four minutes after AstraRCE pointed to the screenshot of the hack, DataRepublican’s X account reposted the image, alongside a caption characterizing NoBlogs as a “platform shared by Antifa and far-left Marxists.”

“The State Department designed them as a Specially Designated Global Terrorist group yesterday,” DataRepublican wrote. “And now … they’re TrumpBlogs.”

Later, DataRepublican’s website published a visualization tool that allows users to view thousands of NoBlogs sites dispersed across a map, naming the tool Trumpblogs, in what appears to be a reference to the hack. Shortly after the tool went online, AstraRCE posted on X to say that the trumpblogs.org domain would now be redirecting to DataRepublican’s site.

The Pentagon has thus far declined to reveal what assignment Pounds or other right-wing media influencers have been given, and it’s unknown if activities like producing the visualization tool would be part of these duties. (The Department of War did not reply to a request for comment.)

“They’re hiring people who have a history of sort of dirt digging,” said David Greene, who serves as legal counsel at the nonprofit Electronic Frontier Foundation. “And they’re digging for information about their enemies, and both trying to identify their enemies and build dossiers on them.”

Holmes, of the Freedom of the Press Foundation, said that determining responsibility for cyberattacks like the one that targeted NoBlogs can be incredibly difficult, especially for smaller organizations.

“Attribution is really hard to prove definitively,” Holmes said. “But, also, there are a lot of imbalances there because the folks at Autistici don’t necessarily have the ability to perform the type of forensics and response that normally leads to successful attribution, or at least not quickly.”

Partners in Posting

DataRepublican and AstraRCE steer clear in their posts of explicitly claiming responsibility for hacks. They frequently, however, offer kudos to fellow travelers. Sometimes they post claims of insider access to information about the hacks or, in a few cases, show information that reveals access to the hacked systems themselves.

In at least one instance, a post by DataRepublican appeared to take direct responsibility for a cyber operation. Writing on X in July, the DataRepublican account claimed credit for a honey-pot scheme against a Nigerian hacker, which DataRepublican said had gone after right-wing media personality Hugh Hewitt.

“It’s true,” DataRepublican wrote, attaching a screenshot of the hackers’ location and IP address. “@astrarce and I honeypotted the Hugh Hewitt hacker.”

Even though DataRepublican and AstraRCE don’t take credit for cyberattacks, the far-right web heaps praise on them for the exploits.

Related

The Feds Want to Unmask Instagram Accounts That Identified Immigration Agents

Seeking to disrupt a database collecting information on U.S. Immigration and Customs Enforcement vehicles, a January hack overrode images of license plates with a meme saying, “ALL YOUR LOGINS, LOCATIONS, PASSWORDS, AND PHONE NUMBERS HAVE BEEN GIVEN TO FBI AND ICE.”

After sharing the image and tagging AstraRCE, DataRepublican was asked by an online fan how it happened. The account replied, “I don’t know, but an unrelated tip is that it’s a bad idea to make your encryption/decryption script readable to the world.”

Several X accounts in the right-wing milieu responded to DataRepublican’s initial post with congratulations. A spokesperson for Turning Point USA said, “Well done, patriots.”

In another case, a Fox News article purported to expose the network of organizers behind protests at ICE’s Delaney Hall detention center in New Jersey. Summarizing its report, Fox News credited, among others, DataRepublican and AstraRCE with infiltrating the group: “The activities of this network have motivated a group of tech sleuths on the X — @DataRepublican, @Astrarce, @bitchuneedsoap and @gunshymartyr — to penetrate these groups, their Signal chats and their operations like a digital Avengers squad.”

In February, following a cyberattack on Siembra NC, a North Carolina-based immigrant rights organization, DataRepublican claimed the group operated two active websites responsible for “doxxing ICE.” Citing AstraRCE, DataRepublican suggested hackers had taken over the nonprofit’s notification system.

“Per @astrarce,” DataRepublican wrote, “the 1000+ users signed up to both sites have received a text alert with the meme attached” — a photo of FBI Director Kash Patel with meme text that says, “FEDS HAVE ALL YOUR DATA.”

Related

“They Want People Scared”: Former Palestine Action Member Blasts Trump’s Terror Sanctions

DataRepublican then posted what appeared to be a screenshot of the organization’s internal notification dashboard, claiming that more than 2,000 users had received push notifications. The notification text visible on the dashboard suggested users received a message urging self-deportation and claimed their personal information had been sent to federal authorities.

During a similar hack of StopICE.net, an ICE tracking website, DataRepublican claimed to have insider knowledge of data from a hack being sent to the authorities.

“I can personally testify,” DataRepublican posted, “that the names, logins, passwords, locations (down to coordinates), and phone numbers of 100K+ users were all sent to FBI, ICE, HSI” — ICE’s Homeland Security Investigations division — “and more.”

Like the cyberattack targeting NoBlogs, in other cases DataRepublican and AstraRCE were also quick to celebrate cyberattacks against leftist groups and post screenshots of the incursions. In many of the cases, the webpages of the sites were vandalized and a message was left behind claiming to have accessed and sent internal data to the federal government.

Claiming Inside Info

The NoBlogs hack became one of the instances where DataRepublican and AstraRCE alluded to both having insider information about the provenance of materials from the hack and their transmittal into the hands of the authorities — all while Pounds was firmly ensconced at the Pentagon.

The episode, in the immediate wake of the NoBlogs hack, was telling. AstraRCE echoed the claim made on the vandalized NoBlog homepage: that the federal government was in possession of the site’s internal user data.

The episode was set off when the A/I collective issued a statement acknowledging the hack and that the “attacker may have had access to the users database.” The collective warned that the hackers could have accessed the “list of email addresses of the people who manage the blogs and their passwords.”

The password information for the user accounts, however, was stored “hashed,” A/I said, rather than in plain text, making it substantially harder to access.

Shortly after A/I published its statement about hashed passwords, AstraRCE posted an image of the statement alongside a caption suggesting that they had direct knowledge of the data that was breached.

“Yep,” AstraRCE wrote. “The government has their emails and dehashed results.”

One minute later, DataRepublican verified AstraRCE’s statement: “Can confirm the authorities have everything that was on the NoBlogs servers.”

The post Pentagon Employee DataRepublican Posts Inside Info About Hacks Against the Far Left appeared first on The Intercept.

Though Flock’s investors and executives insist its surveillance dragnet thwarts the most serious of crimes, it turns out police are also using the company’s technology to go after some of the pettiest possible offenses: curfew violations, loitering, and simply being homeless.

Related

License Plate Surveillance, Courtesy of Your Homeowners Association

When police seek footage from Flock’s more than 120,000 cameras nationwide, they select a category of crime corresponding with the alleged offense they are investigating, such as hit-and-run or burglary. Among the choices is the supposed crime of being suspicious in public, formally categorized by Flock as “Curfew/Loitering/Vagrancy Violations.”

Records obtained from the police department in Plano, Illinois, reveal that law enforcement agencies nationwide conducted more than 1,200 Flock searches while investigating alleged “Curfew/Loitering/Vagrancy Violations.”

As this data set reflects only searches of Flock cameras that extended to Plano’s network, this is assuredly just a tiny fraction of the total number of “Curfew/Loitering/Vagrancy Violations” investigations using Flock nationwide. Records hosted by the website Have I Been Flocked? — which aggregates such logs — show hundreds more of these loitering searches in other states.

The Plano searches were conducted by police departments throughout Illinois, from neighboring states, and from a Georgia town more than 600 miles away. Many queries were license plate searches pertaining specific vehicles; others show police surveilling people on foot for the crime of essentially standing suspiciously.

Plano’s deputy chief of police Gene Morton told The Intercept the town did not install Flock cameras until the end of February. The department shared data from January because it had previously signed an agreement with Flock to search the broader network before its own cameras were operational.

Flock did not respond to a request for comment.

Police in Norcross, Georgia, for example, conducted a smartphone-based search on January 31 with “Curfew/Loitering/Vagrancy Violations” as the stated justification. It tapped into data recorded by 6,098 different Flock camera networks, each consisting of an untold number of separate cameras that surveilled anyone who moved past them.

The search logs obtained by The Intercept do not break out “Curfew/Loitering/Vagrancy Violations” separately, so it is unknown in each case what exactly police were searching for. In some cases police appended extra information; the Tennessee Bureau of Investigation, for example, searched over 17,000 Flock cameras while investigating “Curfew/Loitering/Vagrancy Violations – Suspicious person at planned protest site.”

Related

Philly Cops Admit That They’re Tracking “First Amendment Activity” Critical of AI

When asked about these queries, Tennessee Bureau of Investigation spokesperson Josh DeVine told The Intercept that the searches in fact originated with the state’s Tennessee Fusion Center, which is operated by both the TBI and the Tennessee Department of Safety and Homeland Security. DeVine referred The Intercept’s questions to the Tennessee Department of Safety and Homeland Security, which did not immediately respond to a request for comment.

Flock cameras have also been used to zero in on the homeless: Dave Maass, director of investigations at the Electronic Frontier Foundation, shared Flock search data from the West Palm Beach Police Department with The Intercept that included a query pegged to an “outside agency dropping vagrant off” in the city.

In January, police in LaGrange, Georgia, searched more than 6,000 Flock networks on the basis that someone walked out of a house “with a significant amount of traffic.” Officers from the same agency earlier this year searched more than 15,000 Flock systems because of someone “lingering at a stop sign around midnight last night.”

In the data reviewed by The Intercept, most of these loitering searches didn’t include any corresponding case number, suggesting these queries were not part of broader criminal investigations. Some of the searches reviewed by The Intercept note a “suspicious” person spotted outdoors, without any additional justification for the dragnet.

Jay Stanley — a senior policy analyst at the American Civil Liberties Union’s speech, privacy, and technology project — told The Intercept these searches indicate surveillance mission creep. “This is in line with what we see with these technologies sold on the basis of movie plot scenario, dramatic ‘saving the day.’ And yet when you look under the hood, they’re often used to nickel and dime people for the most prosaic offenses,” he said. “They become part of many departments’ over-policing.”

Loitering and vagrancy laws have an ignoble history across the country as pretextual statutes, many originating in the Jim Crow era, for the targeting of Black Americans guilty of nothing other than existing in public. Some of these laws have been struck down because of the extent to which they criminalize the most mundane behaviors: standing, walking, or simply hanging out. In 1972, the Supreme Court deemed unconstitutional a Jacksonville, Florida, loitering ordinance targeting, among other things, “persons wandering or strolling around from place to place without any lawful purpose or object” and “habitual loafers.”

Stanley pointed out that any guardrails against abuse of Flock’s camera networks is entirely self-enforced by the company, which has an interest in police using it as much as possible, and self-oversight by the police themselves. “We know that police stations don’t seem to be doing this,” he added. “Most police departments are not really paying attention.”

Correction: September 10, 2026, 2:46 p.m. ET
This article was corrected to indicate that cameras in Plano, Illinois, were not searched in each of the 1,200 “Curfew/Loitering/Vagrancy Violations” queries recorded by that town’s police department. Plano provided data about Flock searches starting in January, but the town’s own Flock cameras were not installed until February.

The post Cops Are Using Flock to Spy on People for the Crime of Standing Around appeared first on The Intercept.

The desire for a custom AI tool with “minimal refusal rates” to Pentagon commands was revealed in files released to The Intercept as part of a Freedom of Information Act lawsuit seeking information about the military’s secretive deals with AI companies.

Related

AI Giants Work Hand-in-Hand With the Pentagon, Contracts Reveal

OpenAI, along with rivals Google, xAI, and Anthropic, all agreed in 2025 to develop militarized prototypes of their state-of-the-art AI to assist the armed forces in uses including logistics, intelligence decision-making, and general “warfighting.” Earlier this year, the Pentagon sought to expand the scope of these agreements and deploy them across U.S. classified computer networks, resulting in a high-profile showdown with Anthropic over whether and how the company could restrain military uses of its technology.

The clause seeking “minimal refusal rates” from OpenAI appears in an updated contract — version “P00003” — expanding upon last summer’s prototype deal, worth up to $200 million over the contract’s two-year duration. A separate document, signed by both OpenAI and the government on February 6, indicates that OpenAI agreed on that day to the contents of an expanded version “P00003.”

OpenAI and the Pentagon deny agreeing to such “minimal refusal ” language, claiming that the “P00003” document provided to The Intercept was a draft and not the final version. “OpenAI has never agreed to contract language requiring ‘minimal refusal rates.’ This language does not appear in our executed contract,” said spokesperson Nate Evans.

“The document you received appears to be an earlier draft proposed by the Department before we provided feedback. We rejected that language, the department agreed to remove it, and the final executed agreement does not include it,” Evans said.

Working with Legal Advocates for Safe Science and Technology, The Intercept’s FOIA inquiry specifically sought only final, executed contract documents. The request asked the Pentagon to exclude any draft materials. None of the documents released through the lawsuit is marked as a draft.

When asked to confirm whether the document containing the “minimal refusal rates” clause was in the final agreement, a Department of Justice attorney representing the Pentagon in the lawsuit said it was indeed the signed and executed version of the contract.

Hours later, and following The Intercept’s outreach to OpenAI, however, the Pentagon’s lawyer said to disregard the prior confirmation, stating that the Department of Defense required more time to investigate the matter.

The Intercept was then contacted by Trevor Tiedeman, a special assistant to the under secretary of war for research and engineering, who prior to his Pentagon position worked for President Donald Trump’s reelection campaign. “There might be some stray voltage or wires crossing between whatever FOIA information you may have received versus what actually exists versus what is an executed contract per se,” Tiedeman said in an interview.

He suggested the document containing the “minimal refusal rates” clause may have been a draft copy, but said he was unsure of exactly what the document was, why it was produced to The Intercept pursuant to its FOIA request and lawsuit, or why the Department of Defense abruptly reversed course.

Related

Hegseth’s War Department Slashed Civilian Protection Staff, Brought in AI Assessments

Tiedeman told The Intercept he would provide a full accounting of how the document was erroneously flagged by Pentagon FOIA officers, cleared for release by the Department of Defense, and then confirmed as accurate by the Pentagon’s lawyers. When asked if the Pentagon could share the correct version of the OpenAI contract it ostensibly neglected to release, Tiedeman said he would investigate the matter. He then stopped responding to The Intercept’s inquiries.

OpenAI similarly did not provide the full contract. (In 2024, The Intercept sued OpenAI in federal court over the company’s use of copyrighted articles to train its chatbot ChatGPT. The case is ongoing.)

Days later, the Justice Department lawyer representing the Pentagon further backtracked, stating the document in question “was not the final version of that document,” and that the “correct document” would be shared later, “although I do not have a definitive timeline.”

Department of Defense spokesperson Jacob Bliss later said in a statement “the phrase ‘minimal refusal rates’ does not appear in any active Department of War contract with OpenAI.”

The language indicating the military sought a more pliable version of OpenAI’s technology is found in a section of a contract document describing what OpenAI was obligated to deliver to the Pentagon. These deliverables included “Testing, Evaluation, and Refinement of OpenAI Mission Models” for “national security problem sets.” The document explains “’OpenAI Mission Models’ refer to OpenAI models that are designed for national security use cases and have minimal refusal rates.”

There is no indication in the paperwork about what these “national security problem sets” may entail. Nor is there any description about the kinds of requests the Pentagon hoped OpenAI’s technology would minimally refuse. However, a large language model that would aid in the process of spying on, targeting, and killing people would require a substantial relaxation of safeguards to be useful for any military.

Related

It’s Time to Rein in Lethal AI Drones

Like many of its rival platforms, OpenAI’s flagship LLMs contain built-in guardrails that direct the tool to reject certain queries, typically on the basis of safety. Asking the consumer version of ChatGPT for help prioritizing drone-based airstrike targets, for example, generates this reply: “I can’t provide a prioritization scheme for conducting UAV airstrikes against specific enemy combatants.” OpenAI and its competitors ban the public from using their models for other dangerous applications, like the development of weapons of mass destruction.

Heidy Khlaaf, chief scientist at the AI Now Institute and former systems safety engineer at OpenAI, told The Intercept the notion of national security-specific guardrails is in itself worrying. “Minimal refusal is likely referring to little or no safeguards on the model being used,” Khlaaf said.

No matter the final contract language, experts like Khlaaf are concerned about the role corporate policy is already playing in combat. “It is a worrying development that private corporations are given the power to [make] determinations in warfare that are ultimately state obligations, whether it be for targeting recommendations, or the guardrails deployed to constrain a state’s military use.”

Questions of what limits — if any — tech firms can or should place on battlefield usage were at the center of this year’s public brawl between the Pentagon and Anthropic. The company claims its military deal to expand into classified networks collapsed when the Defense Department refused to place contractual prohibitions against the use of its technology for autonomous weapon systems and domestic surveillance. The Trump administration in turn designated Anthropic a supply-chain risk and moved to ban it from government use (the designation was overturned late last month by a federal judge).

Such tensions didn’t stop OpenAI from quickly cementing its version of the deal. On February 27, OpenAI signed the latest iteration of its Pentagon agreement, permitting the use of its services across the U.S. military’s classified networks. The classified deployment contract update includes the same “Testing, Evaluation, and Refinement of OpenAI Mission Models” header, but its text, unlike the previous version, is redacted entirely.

OpenAI claimed that it secured red lines on autonomous killings and spying against Americans. But the way the deal is drafted ultimately allows for any uses the government deems legal.

The post The Pentagon Asked OpenAI for Artificial Intelligence Designed to Rarely Say No appeared first on The Intercept.

The documents, many heavily redacted, consist of contract paperwork from deals originally signed in July 2025 between the Department of Defense and Anthropic, Google, OpenAI, and xAI, as well as later amendments to those agreements. Through the contracts, each worth up to $200 million, the companies agreed to build prototypes of artificial intelligence tools to “improve military advantage, military utility, or enhance military decision making” across the whole of the armed forces. These prototypes, aimed at applications from “warfighting” and automated decision-making to logistics and military intelligence, would go on to form the basis of the artificial intelligence tools now deployed across the Department of Defense’s classified networks. The document set includes one contract reflecting OpenAI’s classified network deployment; The Intercept is awaiting the government’s production of equivalent documents from the other contractors.

Related

The Pentagon Asked OpenAI for Artificial Intelligence Designed to Rarely Say No

The contract materials show the close and complex partnership between the Pentagon and the AI industry, and detail the extent to which AI firms themselves are expected to influence how their technology is used to augment the U.S. capacity to wage war.

“This appears to be the first time the extent of collaboration between frontier AI labs and the Pentagon is spelled out in explicit terms,” Sophia Goodfriend, a University of Cambridge research fellow and non-resident fellow at the Harvard Kennedy School’s Middle East Initiative studying the impact of big data and machine learning on military conflict, told The Intercept. “What’s particularly notable is the terms by which engineers are working in lockstep with the department of war to engineer AI systems for surveillance, targeting, and killing.”

Official announcements following the 2025 deals were light on details and heavy on platitudes. “We must equip our warfighters with 21st-century technology in order to defeat 21st-century threats,” explained Pentagon spokesperson Sean Parnell at the time.

Google was similarly cagey: “These advanced AI solutions will enable the DoD to effectively address defense challenges and scale the adoption of agentic AI across enterprise systems to drive innovation and efficiency with agile, proven technology,” the company said.

The contract documents for the prototyping project paint a clearer picture, revealing deals that involve far more than a simple sale of software.

An “information sharing” provision in all four agreements, for instance, creates a two-way exchange of data and expertise. The companies agreed not only to provide access to their best large language models, but also to “inform DOD and CDAO [Chief Digital and Artificial Intelligence Office] strategy on frontier AI and AI,” as phrased in the agreement with Google.

Under the agreements, the AI labs were approved to receive an array of sensitive information, which could include “benchmarks datasets for DoD use cases, appropriate briefings on DoD operational missions and threats to inform development / release of technologies, their application to DoD operational problems, or DoD plans and strategies for future AI adoption.”

All four companies would in turn provide the Pentagon “feedback on DoD strategies for frontier AI adoption to ensure that AI capabilities can be delivered and scaled to meet the warfighter’s needs.” The companies also pledged to offer briefings on corporate inner workings, including “frontier AI model performance, case studies on current or likely future frontier AI applications, or projections of future trends in frontier AI maturation.”

Related

Anthropic Says We Must Stop Authoritarian AI. But What About Its Authoritarian Investors?

The labs would sometimes be the ones offering intelligence briefings on foreign actors, with Google tapped to inform the Pentagon on “AI tactics and techniques of adversaries” of the United States. These briefings would be conducted in both unclassified and classified settings.

The collaboration also entails Big Tech providing training and educational seminars directly to the military, including “Presentations and briefings at technical exchange meetings on relevant topics, including frontier AI model performance and case studies of frontier AI applications.” One document notes that Google, for example, would lead multiple seminars on “responsible AI” deployment.” The documents state engineers and policy experts from all four contractors will conduct “tabletop exercises,” gamified simulations of real-world scenarios, with the Pentagon.

This back and forth between the military and private sector would, the Pentagon hoped, allow for “iterative refinement of frontier AI models to address real-world challenges in areas such as intelligence analysis, cybersecurity, and autonomous systems,” according to contract language with all four companies.

The AI companies were contractually scheduled to provide “risk forecasting, and threat ideation exercises” — predictions of pitfalls and dangers their own products might pose in the future. “Frontier AI labs have advanced risk forecasting tradecraft and a clear vision of the next wave of frontier AI technological developments,” the contracts read. “Frontier AI companies can help DoD gain a better understanding of the risks that might be associated with subsequent near-term frontier AI models or features, to avoid strategic surprise and to mitigate the associated risks.”

Heidy Khlaaf, chief scientist at the AI Now Institute and former safety engineer at OpenAI, told The Intercept this is a dubious claim with dangerous implications. Pointing to recent disclosures by OpenAI and Anthropic that semi-autonomous large language models broke into computer networks owned by other companies during testing, Khlaaf questioned their fitness to help build guardrails in the first place. “This is a very concerning development,” she said.

According to Khlaaf, the “risk forecasts” so far offered by frontier AI labs have been light on evidence and tend to “emphasize self-beneficial skewed risks such as a purported AI arms race and speculative ‘existential’ risks,” like self-aggrandizing science fiction-style “Terminator” scenarios.

The public would be better served, she said, with a reliance on independent assessment of the risks these companies present, not the word of the companies themselves. Trusting companies to self-report the risks of their own products constitutes both a conflict of interest, Khlaaf added, and a “subversion of democratic processes when AI labs are allowed to take over the arbitration of risk determinations with life-or-death consequences.”

As the prototyping effort progressed, the contract documents show the scope of the project expanded. New versions of the contract show additional deliverables included in the list of what the Pentagon was purchasing. The government redacted descriptions of these new deliverables on the basis of military secrecy and corporate confidentiality.

It was during the negotiation of these subsequent amendments when Anthropic and the Pentagon had their falling out. Anthropic refused to sign a follow-up deal to allow deployment of its technology on classified military networks without contractual prohibitions against use for domestic spying and autonomous weapons, leading Secretary of Defense Pete Hegseth to designate the firm a “supply chain risk” in March and ban its services from government use, a decision overturned by a federal judge last month.

Related

OpenAI on Surveillance and Autonomous Killings: You’re Going to Have to Trust Us

The Department of Defense quickly reached follow-up agreements with Anthropic’s main competitors — Google, OpenAI, and xAI — to apply the tools they had been developing since the summer of 2025 to classified military networks. OpenAI’s deal, described by CEO Sam Altman as “definitely rushed,” was reportedly finalized on February 27. The following day, the company published a blog post defending “Our agreement with the Department of War” and insisting there were clauses prohibiting its technology’s use for domestic surveillance.

An amended contract document obtained by The Intercept through its lawsuit dated February 27 contains new language, including a clause noting that OpenAI will embed company engineers within the military to aid U.S. Government employees in implanting and utilizing its AI system. These workers, the document notes, “can be deployed to warfighting support settings including, but not limited to, combatant commands, service components, and theatre components.” The February 27 amendments also include a section on “System Oversight” is redacted in its entirety.

None of the documents released to The Intercept pertaining to xAI or Google’s equivalent agreements contain any such oversight section.

Two documents indicate the Department of Defense asked OpenAI to minimize the extent to which its large language model would refuse to do what was requested by the military. The Department of Defense and OpenAI said those documents were draft materials incorrectly provided to The Intercept, and that the final contract included no such provision.

(In 2024, The Intercept sued OpenAI in federal court over the company’s use of copyrighted articles to train its chatbot ChatGPT. The case is ongoing.)

Related

OpenAI Quietly Deletes Ban on Using ChatGPT for “Military and Warfare”

The contractual coziness with the world’s most powerful military indicates just how far the artificial intelligence industry has strayed from its earlier public commitments to safety and global prosperity. OpenAI, originally founded as a nonprofit “to ensure that artificial general intelligence benefits all of humanity,” up until two years ago explicitly banned “military and warfare” use in its terms of service. The following year Google reversed a similar self-imposed ban on using its state-of-the-art AI services for surveillance or “technologies whose principal purpose or implementation is to cause or directly facilitate injury to people.”

Both companies are now eagerly landing slices of the nearly trillion-dollar Pentagon budget, and for opportunities to augment the killing power of a military that has recently bombed schoolgirls in Iran and civilian vessels in the Caribbean.

Goodfriend, the researcher, said the contracts illustrate the illusory nature of values statements published by these labs. “In recent months companies like Anthropic and Open AI have maintained ‘firm red lines’ when it comes to developing autonomous weapons systems and mass surveillance tools” long after they signed contracts to build “autonomous systems” for the military, she said.

“All this indicates these labs and their personnel have been far more closely integrated into DOD AI development than previously understood and casts some doubt on the supposed contractual limits they have intended to place on the DoD’s use of these systems,” Goodfriend said.

Anthropic, Google, and xAI did not provide comment or answer questions about their respective contracts. In a statement to The Intercept, OpenAI spokesperson Nate Evans said “Our tools are built to support legitimate national security work — including vital cyber defense efforts — while refusing uses that cross the red lines we publicly announced. These restrictions are also reflected in our signed agreement: no use of OpenAI technology for mass domestic surveillance, to direct autonomous weapons systems, or for high-stakes automated decisions. We will continue to work closely with the Department to make sure our AI tools are deployed responsibly and safely.”

Related

The Dawn of a New Cold War

On the long timeline of American warfare, these artificial intelligence tools are in their infancy. The Pentagon’s use of artificial intelligence in the process of killing, however, is not a hypothetical. In March, the Wall Street Journal reported that despite Hegseth’s ban, U.S. Central Command used Anthropic’s large language model services to execute its bombardment of Iran, including for “target identification.”

Despite the intimacy of technology firms and military using it to pinpoint airstrikes, it’s unclear what oversight, if any, these companies have over their own weaponized software once it is in the Pentagon’s hands. When asked if his company’s tools were used in an American airstrike on the first day of the Iran war that killed 120 Iranian schoolchildren, Anthropic CEO Dario Amodei told Bloomberg News he simply did not know. An August 24 report from the New York Times detailed the use of fully autonomous drones guided by machine learning software running on chips made by Nvidia, whose processors have helped make the explosive growth of Anthropic and OpenAI possible. The full extent of the U.S. tech sector’s involvement in killing abroad remains obscured behind trade secrecy and national security considerations.

“The same technology we saw autonomously hacking websites on the open internet weeks ago is being used to inform DoD decision-making,” said Vivian Dong of Legal Advocates for Safe Science and Technology. “The public deserves insight into how and under what constraints the Department is using this technology.”

The post AI Giants Work Hand-in-Hand With the Pentagon, Contracts Reveal appeared first on The Intercept.

Since the beginning of 2025, Flock has given 700 percent more to the partisan groups representing mayors, attorneys general, and governors than in the previous three years. The giving has gone to Democratic and Republican groups alike.

Donating to those groups is a time-worn strategy for companies under scrutiny. The gifts can sometimes reap face time with elected officials. Two major Flock competitors in the license-plate reader industry, Axon and Motorola Solutions, have also donated widely.

One group that received money from Flock, the Democratic Mayors of America, defended its decision to accept $30,000. A spokesperson said reports of Flock misuse are “deeply unsettling,” while arguing that the donations gave Flock no influence over local mayors.

Critics of money in politics have long raised concerns that such donations create, at a minimum, the appearance of a conflict of interest. A surveillance skeptic said the groups could face blowback.

“There’s a reputational risk for them accepting donations from Flock at this time,” said Torin Monahan, a professor of communication at the University of North Carolina at Chapel Hill.

Flock did not respond to a request for comment.

Over the past year, Flock has faced intense pushback from activists on the left and right over perceived civil liberties abuses. Flock was forced to admit that immigration officials had access to its data, despite earlier denials in the middle of the Trump administration’s immigration crackdown last year. Dozens of police officers have also been accused of abusing Flock’s database to track people, often former romantic partners.

Related

License Plate Surveillance, Courtesy of Your Homeowners Association

Flock was founded in 2017, but it was only five years after that that it began doling out money to what are known as 527 groups, according to records compiled by the nonprofit news outlet ProPublica. The so-called 527 groups can raise unlimited funds directly from corporations and spend that money on key political races.

Flock’s giving began at a relatively meager rate: From 2022 to 2024, the company gave $20,000 to the Democratic Mayors Association and $10,000 to its Republican counterpart, the Community Leaders of America.

Its spending has skyrocketed since then. Since the beginning of 2025, the company donated a total of $240,000 to a longer list of groups that includes the Democratic and Republican Governors Associations and the Democratic and Republican Attorneys General Associations.

Flock’s pattern of donating to 527 groups was identified by a widely cited critic of the company, Jason Hunyar, who asked on his blog what the company was receiving in exchange for those donations.

Influence Investments

The donations came as Flock was hiring an army of lobbyists. It has spent more than $2 million on state and federal lobbying since the start of 2025, according to a recent report from the transparency group OpenSecrets.

Related

Before Being Hacked, Border Surveillance Firm Lobbied to Downplay Security and Privacy Concerns About Its Technology

The rise of that multi-pronged influence campaign coincided with intense pushback to the company’s practices. Several members of Congress have proposed either outright bans or stricter regulations on Flock and other license-plate readers.

Much of the activism against Flock thus far has come at the local level. Scores of cities and other local governments have rejected or canceled Flock contracts or deactivated the company’s cameras since the start of 2026, according to the advocacy group DeFlock.

The suspicion crosses party lines. In Knox County, Tennessee, Republican Mayor Glenn Jacobs — the former professional wrestler Kane — recently led a campaign to cancel the company’s contract there. In Arizona, Democratic Attorney General Kris Mayes announced that she was launching a review of Flock and other camera networks.

Flock is not the only license-plate reader company that has routed money to political groups. Motorola Solutions, which counts cameras as one product among many in its portfolio, has donated more than $986,000 since the start of 2023, according to public records.

Related

“Duped”: How One Atlanta Cop Secretly Shilled for Police Tech

Axon, which also hawks Tasers and body-worn cameras, has donated $225,000 to Democratic and Republican attorney general groups over the same period. (Motorola Solutions and Axon did not respond to requests for comment.)

Making Connections

Corporate donations to 527 groups long been controversial. Companies can use the groups to build connections with the same elected officials making decisions about whether to purchase their products and services.

More than a decade ago, the New York Times documented how corporations facing lawsuits used donations to the Republican and Democratic attorney general groups to cozy up to decision-makers.

On its website, the Democratic Mayors Association touts a program that connects mayors “with leaders from the private sector, organized labor, advocacy groups, and partner organizations to build relationships, share ideas, and take on the most pressing challenges facing America’s cities.”

That group defended its decision to accept donations from Flock.

“No donation made to the Democratic Mayors Association guarantees any form of influence over the mayors, no matter who the donor is,” said a spokesperson. “The reporting we’ve seen about police departments across the country misusing Flock cameras and the data is deeply unsettling, which is why mayors across the country have taken action to implement strict guardrails or ban them altogether. Ultimately, Democratic mayors have presided over a historic drop in crime and have proven they know how to keep their communities safe. Each mayor makes their own decisions about what is best for their cities, and we trust their judgement.”

The Democratic Governors Association also defended its donations from Flock and other corporations.

“The DGA does not take policy positions as an organization,” said a spokesperson for the group. “Every dollar sent to the DGA helps us elect more Democratic governors who make their own policy decisions about how to make life better in their states and stand up to Donald Trump.”

Other groups that have received money from one or more of three leading companies in the license-plate reader world — Axon, Flock, and Motorola Solutions — did not respond to requests for comment.

The post Flock Donations to Local Politicians’ Groups Skyrocket as Protests Against AI Cameras Mount appeared first on The Intercept.

Created by tech giant Palantir, a CIA-backed firm known for data mining and using artificial intelligence for everything from warfighting to “predictive policing,” the ELITE app helps ICE target immigrants and issues a “confidence score” for how accurate the information on a target is. The app analyzes data from government agencies and other sources, producing dossiers on targets, compiling them into lists, and mapping their locations.

The debut of ELITE — Enhanced Lead Identification and Targeting — came just as ICE added over 12,000 employees through an expedited hiring surge, placing a badge and a gun in the hands of new recruits in record time, thanks to cuts to ICE’s training curriculum.

Thousands of the new ICE hires, then, were able to hit the streets without cleared background checks, according to reporting by Reuters in early 2026.

The hiring spree at ICE’s Enforcement and Removal Operations division, coupled with the powerful new handheld technology of the ELITE app, created potential civil liberties nightmares and threats to national security, according to ICE whistleblowers, former Department of Homeland Security officials, and legal experts who spoke to The Intercept.

“No one should have been hired by DHS before background reviews were completed. This never happened in previous administrations,” said Mary Ellen Callahan, former chief of staff to the deputy secretary of Homeland Security. “Furthermore, giving the newly hired ICE ERO staff access to sensitive databases such as ELITE threatens public safety by allowing non-vetted teams to access and potentially misuse data.”

The ELITE app, Callahan said, did not go through the usual process government tools undergo to prevent abuses and ensure privacy rights.

“It’s one of the first applications they give new officers so they can start working,” said one ICE deportation officer who came through the hiring surge. “You can access it without full background clearance.”

Related

Trump’s National Killer Police Force

“Some of these guys should’ve never been hired, let alone have access to the information on ELITE,” the officer, who asked for anonymity to protect their livelihood, told The Intercept.

Every ICE deportation officer is issued a phone, where the ELITE app comes standard, according to five ICE officials who spoke to The Intercept. (Neither the Department of Homeland Security nor Palantir responded to requests for comment.)

The five ICE officials, who each work in different regions across the country, said deportation officers without completed background checks continue to work in all their jurisdictions, some of whom were hired nearly a year ago.

No Privacy Assessment

ELITE’s user guide identifies a dozen distinct sources that the app draws from. The databases include sensitive data sets like addresses, criminal histories, immigration records, court rulings, and other personal information about both immigrants and U.S. citizens.

Much of that information would constitute what the government calls PII, or personally identifiable information. According to a Department of Homeland Security document that catalogs the agency’s use of artificial intelligence, the ELITE app does involve PII.

A 2002 law, however, says that government agencies must conduct and publish the findings of a “privacy impact assessment” for “all new or substantially changed technology that collects, maintains, or disseminates personally identifiable information (PII),” according to a Justice Department summary of the statute.

The processes are designed to assess how technology affects personal privacy and outline what safeguards are in place to ensure accountability if systems are misused.

Yet the Department of Homeland Security privacy office hasn’t published an assessment for ELITE, according to Callahan, the former Homeland Security official, who also served as the department’s chief privacy officer.

“The DHS privacy office should have worked with ICE to publish a privacy impact assessment,” said Callahan, noting a general lack of productivity by the privacy office. “In fact, the DHS privacy office has only published one PIA all year. In comparison, I published 228 PIAs in my three-and-a-half years as privacy officer.”

The failure to conduct a privacy impact assessment has a profound impact on transparency and accountability, the very aspects of privacy policy the E-Government Act of 2002 was intended to regulate.

Related

Trump Wants to Put You in a Massive, Secret Government Database

“This is not a technicality,” said Clare Garvie, a privacy lawyer and deputy director of technology policy at the New York University Law School’s Policing Project. “PIAs are an essential component of transparency and privacy governance, they are how the public understands what information is collected about us, how long it is kept, who can use it, and what remedies exist if it is wrong or subject to misuse.”

“If ICE can’t fulfill this prerequisite,” Garvie said, “I lack confidence in its ability to use the tool in a lawful, privacy-protective manner in the field.”

The Department of Homeland Security site for cataloging its uses of AI does list a privacy impact assessment on its page for ELITE, but the assessment is not actually for the ELITE app. Instead, the link goes to a 2019 privacy assessment for ICE’s Enforcement Integrated Database.

The Enforcement Integrated Database is just one of the systems ELITE incorporates into its searches, meaning the PIA only scratches the surface of what personally identifying information appears in ELITE and how the app organizes the data.

Related

LexisNexis Is Selling Your Personal Data to ICE So It Can Try to Predict Crimes

“The document DHS points to as fulfilling this obligation was updated in 2019, years before it acquired ELITE,” said Garvie. “This PIA completely fails to describe the app’s data sources, its AI analysis, or the inferences it produces that are used to focus immigration investigations and removal operations.”

Public concerns surrounding the incorporation of AI surveillance in law enforcement — including ICE’s new AI initiative to combat doxing of its staff — are reaching a boiling point in the U.S. today, leaving some ICE officials concerned.

“I know I sound crazy, but this is 1984 shit,” said another ICE official familiar with ELITE who requested anonymity to protect their job. “ELITE is used on illegal immigrants right now but can easily be used on Americans.”

The post ICE Gave New Hires Access to Restricted Info on Palantir App Before They Passed Background Checks appeared first on The Intercept.

In a federal court filing in New York, the Justice Department argued that the U.S. has a vested interest in OpenAI having access to a vast array of intellectual property. The Trump administration warned that restricting OpenAI’s access to the data could hamper scientific progress and harm the country’s dominance in the race for artificial intelligence.

“The United States has a strong interest in continuing to develop a robust and competitive artificial intelligence industry that sets the standard for the practice and procedure of AI use globally,” said the filing, which was signed by associate attorneys general Stanley Woodward Jr. and Brett Shumate, along with the attorney general’s senior counsel Michael Weisbuch.

Matt Topic, the litigating attorney representing The Intercept in its lawsuit, warned that the Trump administration’s position was out of touch and a threat to the financial survival of media outlets.

“This is pretty reflective of the administration’s pro-AI policies, which are out of step with the public skepticism of AI and out of step with our long tradition in this country of protecting human expression,” said Topic. “If the administration’s position was accepted, it would result in an unprecedented, uncompensated transfer of IP rights from news organizations to tech companies.”

The Intercept originally filed its lawsuit against OpenAI and its partner Microsoft in February 2024, arguing that OpenAI’s use of work reported and written by its journalists constituted a breach of copyright law. In November 2024, the judge then presiding over the case dismissed all of The Intercept’s claims against Microsoft, but allowed its claim against OpenAI to proceed under the Digital Millennium Copyright Act.

The Intercept’s case was later consolidated under U.S. District Judge Sidney Stein alongside a number of other media plaintiffs, including the New York Times, Tribune Media Company, and Ziff Davis. 

OpenAI has argued that its practice of hoovering up staggering amounts of copyright-protected materials to feed its large language models — the underlying technology powering generative AI tools like ChatGPT — falls under the domain of “fair use,” a doctrine protecting the unlicensed use of copyrighted material under certain circumstances.

In its filing, the Justice Department supported OpenAI’s fair-use defense, accusing the Times and other plaintiffs of taking an overly narrow definition of the doctrine. OpenAI’s use of the material, the government argued, met the criteria for “transformative” use, one of the ways in which “fair use” comes into play.

“In this litigation, the New York Times seeks to narrow fair-use doctrine to exclude the training of OpenAI’s large language models (LLMs). That result would be inconsistent with basic copyright law principles and severely hamper ‘the Progress of Science and useful Arts,’” the filing says. “An erroneous fair use ruling would hamper competition in the market for LLMs, because only the largest technology companies might have the capital necessary to pay licensing fees.”

Related

OpenAI on Surveillance and Autonomous Killings: You’re Going to Have to Trust Us

The filing is the latest move from the Trump administration to protect the interests of U.S.-based artificial intelligence companies amid an AI arms race with companies based in other countries, most notably China.

In December 2025, Trump signed an executive order aimed at creating a federal standard, applicable in all 50 states, to “ensure that the United States wins the AI race, as we must.”

Topic, The Intercept’s attorney, said this position — combined with the fact that the Justice Department has not been privy to any discovery in the case — shows that the Trump administration is operating from a fixed position in support of tech companies, rather than arguing the merits of the case.

“What they’re saying isn’t based on any understanding of actual facts,” Topic said. “It just reflects what Microsoft’s and OpenAI’s PR machines have been putting out into the public.”

The post Trump Admin Tells Court: Let OpenAI Rip Off The Intercept’s Articles appeared first on The Intercept.

The memo, signed by Acting ICE Director David Venturella and leaked to The Intercept by ICE employees, identifies a company called ZeroFox as the tech firm behind the initiative. The communication describes ZeroFox as “a modernized protective technology solution designed to strengthen safeguards for our most valuable resource: You and your families.”

The Department of Homeland Security, ICE’s parent agency, quietly entered a $15 million contract with ZeroFox in July for software licenses to support ICE operations and investigations, according to public procurement records. The records indicate that ZeroFox will provide services for the Office of Intelligence at ICE’s Homeland Security Investigations division.

ZeroFox develops AI-driven software aimed at stopping data leaks across social media and the dark web, according to its website. “ZeroFox never stops finding, mapping, and monitoring your digital presence,” the company vows. It’s not clear from the contract or the internal email how, precisely, the new initiative will function, and neither ICE nor ZeroFox responded to requests for comment.

“It sounds like inside surveillance,” said an ICE official who spoke to The Intercept on condition of anonymity for fear of retaliation.

Related

FBI Hired Social Media Surveillance Firm That Labeled Black Lives Matter Organizers “Threat Actors”

ZeroFox has used its tech in the past to identify and categorize people engaged in political activism, including during public unrest in Baltimore after the 2015 police killing of Freddie Gray and the attack on the U.S. Capitol on January 6, 2021.

The ICE official expressed concern that ZeroFox tech could now be used to weed out whistleblowers and critics of the Trump administration.

“If things go sideways they may just throw all your info out there, make you out to be the problem or dox you,” the official said. “I don’t trust it.”

According to Richard Forno, associate director of the cybersecurity institute at the University of Maryland, Baltimore County, it’s not shocking that ICE would contract with AI companies that “may have checkered histories.”

“It’s plausible that this tech could be used to seek out whistleblowers,” Forno told The Intercept. “For example, if ICE internal affairs or leadership can seed information that gets leaked, and then that information is then passed along to a reporter, you’re kind of setting up a honey pot situation to figure out who the leakers are.” He pointed out that the White House has already set a precedent for targeting critics as “terrorists” or “extremists” on fairly broad terms.

In 2015, ZeroFox offered its services to Baltimore officials for free, sending Baltimore Police a list of “threat actors” identified by its social-media monitoring system, which classified peaceful BLM protest organizers as physical threats. (ZeroFox’s co-founders told the press at the time that the classifications were likely based on their social media follower counts, and that they were labeled “physical” because they were “acting in real life” rather than online.)

After the January 6 attack, the FBI struggled to identify online extremist activity using its recently acquired ZeroFox technology, which the bureau’s agents referred to with “a derisive nickname,” according to the Washington Post, that involved “replacing the ‘Fox’ with a similar-sounding expletive.”

Related

A Redditor Criticized ICE. Trump Is Trying to Unmask Them by Dragging the Company to a Secret Grand Jury.

ZeroFox’s relationship with ICE, according to the leaked memo, seems to be focused on detecting information online that could identify or “dox” ICE employees. It’s understandable for a person to want to limit the personal information published about them online, said Sophia Cope, a senior staff attorney at the digital rights group Electronic Frontier Foundation.

“However,” she said, “if ICE is instituting this program to shield officers and the agency itself from public accountability, especially when information is already publicly available, then that’s a serious concern and flies in the face of our democratic values.”

ICE officials who spoke to The Intercept said some ICE personnel already use third-party services to keep their information private.

“I frankly don’t trust the feds with that,” said a second ICE official, who spoke to The Intercept under anonymity to protect his livelihood. “I’m not really keen on giving my PII” — personally identifiable information — “to a company that doesn’t have a solid track record with government contracts.”

ZeroFox is owned by private equity firm Haveli Investments, which bought the company for roughly $350 million in 2024. Bryan Ware, a former assistant director for cybersecurity for the Department of Homeland Security during the first Trump administration, currently serves as an executive at ZeroFox.

The post ICE Is Paying a Controversial AI Firm to Hide the Identities of Agents appeared first on The Intercept.